最新刊期
卷 57 ， 期 1 ， 2025
- “在网络安全领域，专家建立了面向暗网抑制的普适性安全理论体系，为精准治理暗网犯罪提供解决方案。”摘要：Significance In recent years, anonymous networks and their underlying darknet have become vital tools for transmitting sensitive information, conducting cyberattacks, and engaging in cybercrime due to their strong concealment, high anonymity, and resistance to traceability. These characteristics pose serious threats to national security and social stability. This project researches a universal security theory for darknet suppression to address the challenges of darknet governance, such as difficulties in identifying concealed communication behaviors, mapping dynamic network topologies, and disguising trap node deployments.Progress The main content includes: 1) Establishing a collaborative quantitative theoretical framework focused on darknet traffic differences and behavioral commonalities. This involves proposing heterogeneous darknet universal characteristics, differentiated element representations, unified security quantification, and ecological vulnerability graph construction theories. These approaches address the challenge of quantifying darknet suppressibility, which remains complicated by diverse network structures and dynamic communication behaviors. 2) Proposing a real-time lightweight traffic detection method based on solving convex optimization problems. This involves constructing a small flow sampling model based on self-similarity associations and a darknet traffic identification and service classification model using Gaussian kernel functions and multimodal optimization. This method enables precise, real-time identification and classification of darknet traffic. 3) Introducing a multi-network full-time domain connection prediction and relationship mapping method based on behavioral invariance. This approach represents cross-point connections and filters out irrelevant connections in dynamic networks to predict multi-network full-time domain connections and map relationships, achieving multi-point global associations of darknet connections under local observation conditions. 4) Proposing a trap node deployment and tracing optimization method for darknet connections based on local observations, enabling tracking and tracing of the darknet under conditions of partially controllable nodes. 5) Developing a real-time traffic detection and tracing demonstration system for real-world darknet scenarios, which law enforcement agencies implement to achieve precise governance of darknet-related crimes.Conclusions and Prospects This project significantly contributes to darknet governance by developing a quantitative framework for analyzing and managing darknet traffic. The proposed real-time lightweight traffic detection method enhances law enforcement’s ability to identify and classify darknet activities. In addition, these methods for predicting multi-network connections and optimizing trap node deployment improve tracking capabilities in complex environments. Future work focuses on refining these methodologies and exploring additional dimensions of darknet behavior to strengthen efforts in combating illicit online activities, generating meaningful social and economic benefits.关键词：darknet governance;traffic detection;behavior recognition;security quantification;trap node deployment1245|688|0更新时间：2025-01-17
FOCUS ON STATE KEY RESEARCH DEVELOPMENT PROGRAM OF CHINA
- “在天然山区河流推移质监测领域，专家全面整理了监测技术研究进展，指出了未来研究重点和亟待解决的科技问题。”摘要：Significance Accurately monitoring bedload transport in mountain rivers is critical for understanding sediment dynamics, predicting geomorphological changes, and informing sustainable river management practices. Current techniques face considerable limitations, particularly in the context of natural mountain rivers. These regions pose unique challenges due to high flow rates, complex sediment compositions, and rapidly changing environmental conditions. Global climate change further complicates the situation, significantly altering river hydrology and sediment transport.Progress This study provides an exhaustive review of the current state of bedload monitoring technologies, both domestically and internationally. Direct measurement techniques are thoroughly examined, including instrument-based and pit-slot methods. Instrument-based methods, such as sediment samplers and sediment traps, experience substantial advancements in accuracy, efficiency, and data collection capabilities. However, these methods face significant limitations, particularly regarding their applicability during high flood conditions. Issues such as equipment damage, safety concerns, and the inability to capture sporadic and highly concentrated bedload events present considerable challenges. Pit-slot methods involve the excavation of pits or slots in the river bed to directly measure sediment load. Although these methods yield accurate measurements, they are labor-intensive, disruptive to the natural river environment, and often impractical in high-energy mountain rivers. Indirect measurement methods, capable of long-term, continuous recording of signals related to bed changes and bedload transport, are extensively reviewed. These methods are primarily divided into active and passive methods. Active methods, such as acoustic Doppler velocimetry and laser Doppler anemometry, use externally generated signals to measure bedload transport. Although these methods provide the advantage of non-invasive measurements, they are affected by factors such as water turbidity, signal scattering, and interference from suspended sediment. Passive methods are further categorized into contact and non-contact types. Contact methods, such as bedload impact plates, columns, and pipes, rely on the impact of moving sediment to generate signals. Non-contact methods, including seismometers and hydrophones, detect the vibrations or sounds produced by bedload movement. These methods enable continuous, real-time monitoring and provide valuable insights into sediment transport dynamics. However, the signals recorded by these sensors are often complex and influenced by various environmental factors. This complexity creates significant challenges in signal interpretation and quantitative calibration, making this a research hotspot with ongoing efforts to develop more effective signal processing and calibration techniques. The study concludes by emphasizing the need to construct prototype observation stations, integrate multi-source signal monitoring technologies, and develop watershed bedload monitoring equipment systems. These aspects require immediate resolution of technological challenges to enhance hydro-sediment monitoring capabilities. Understanding the spatiotemporal distribution characteristics of bedload transport in mountain rivers and its relationship with geological environments and climatic conditions in the basin remains a crucial research priority. Conclusions and Prospects This understanding further raises the development of bedload measurement technology and computational theoretical methods, providing fundamental data support for studying the global sediment source-sink system. It also provides significant references for water resource environmental management, sustainable development planning in mountain basins, disaster prevention and mitigation, and major national engineering construction in China. Despite extensive research and progress, the complexity of bedload monitoring issues presents considerable exploration space based on existing advancements and results. The key and urgent technical problems include constructing prototype observation stations, integrating multi-source signal monitoring technologies, and building watershed bedload monitoring equipment systems. These challenges emphasize the importance of continued research and development in this field to improve the understanding of sediment dynamics in mountain rivers and to contribute to more sustainable river management practices.关键词：mountain river;Bedload transport;Monitoring technologies;Research advances373|616|0更新时间：2025-01-17
-
Predictive Method for the Evolution of Flow Velocity in an Open Channel with a Floating Vegetation AI导读
“在河道水环境治理领域，专家基于指数衰减函数建立了漂浮植被区水流流速沿程分布预测方法，为生态修复提供技术支撑。”摘要：Floating canopies are often distributed in open channels, altering the flow velocity evolution and affecting the nutrient removal efficiency of floating vegetation. The flow in an open channel with a floating vegetation canopy is vertically divided into the vegetated region and the bare channel. As the flow enters a floating canopy, the flow along the canopy’s leading edge adjusts, and the in-canopy velocity continuously diminishes over the interior adjustment distance, XD, due to vegetation drag. Beyond XD (x > XD), the flow becomes fully developed, and the velocity remains constant. Inside a floating canopy, the interior adjustment distance, XD, is defined as the distance from the canopy’s leading edge to the position where the flow is fully developed. A model based on exponential decay and the flow continuity equation is proposed to predict the longitudinal profiles of streamwise velocities within and below a floating canopy. The proposed model does not include a calibration parameter, and all input parameters can be obtained from flow and vegetation conditions, which is advantageous for its application. Twelve groups of experimental data from various sources are utilized to verify the proposed model. The model accurately predicts the velocity evolution within and below the floating canopy of vegetation. This study proposes a predictive model for the longitudinal profiles of streamwise velocities in an open channel with a floating canopy, providing a theoretical basis for river ecological restoration and water resource management.关键词：floating vegetation;flow velocity distribution;prediction model;nutrient removal255|382|0更新时间：2025-01-17 - “最新研究揭示了植被对弯曲分汊河道冲刷与再造过程中水流和泥沙输移特性的影响，为河道整治和生态修复提供理论支持。”摘要：Rivers, particularly meandering branching channels, play an essential role in human production, life, and the development of civilization. Their unique topography and complex flow and sediment transport characteristics significantly impact river stability, flood control, navigation, and economic development along the coast. In meandering branching channels, the widening of diversion areas and the influence of bend circulation cause sediment to accumulate on the convex bank, forming shallow banks that raise vegetation growth. This process triggers solid bank formation, reduces flow velocity, raises water levels, and induces other effects. However, limited research has addressed meandering branching channels influenced by vegetation, and resolving the complex relationship between flow-sediment dynamics in these channels remains an urgent challenge. This study conducts generalized physical experiments to investigate the unsteady flow of meandering branching channels under various inflow conditions and vegetation densities. Flow and sediment data are collected using techniques such as ultrasonic distance measurement, video stream gauging, and instantaneous sediment transport rate measurement. The study analyzes the characteristics of water flow and sediment transport during the erosion and reconstruction processes of meandering branching channels influenced by vegetation. Key aspects include changes in branch diversion ratios, averaged bedload transport rates per unit width, and variations in the characteristic parameters of transported bedload. The experimental results showed that under unsteady flow conditions, the diversion ratio of the left branch channel is significantly affected by vegetation and tends to increase with higher vegetation density. In contrast, the diversion ratio in the right branch is more sensitive to the presence or absence of vegetation cover, while vegetation cover density has a lesser effect. Regarding bedload transport, vegetation in the branch channel increases the sediment transport rate and sediment grading. As vegetation density in the right branch increases, the peak bedload transport rate rises, and fluctuations in bedload transport become more pronounced, particularly at higher flow rates. In addition, changes in quantified parameters of channel interactions reveal that the influence between channels during the reconstruction process intensifies significantly with the increased vegetation density in the branch channel. These findings provide theoretical support for channel regulation and river ecological restoration.关键词：meander branch channel;vegetation effects;flow characteristics;bedload transport;unsteady flow214|288|0更新时间：2025-01-17
- “黄河多沙粗沙区重要支流来沙变化研究取得进展，分析了产沙输沙变化的多时空属性特征，为调控治理黄河水沙灾害提供思路。”摘要：Objective The significant accumulation of silt constitutes a critical factor contributing to water and sediment disasters in the Yellow River. In recent years, the runoff and sediment transport of the Yellow River and its primary tributaries have decreased significantly. Although extensive research has been conducted, the multi-temporal and spatial scale characteristics of sediment production and transport remain unclear.Methods This study examines the Huangfu River and Kuye River tributaries in the high coarse sediment area of the Loess Plateau to analyze the multi-temporal and spatial characteristics of sediment production and transport. The analysis integrates modern hydrological observation data and detrital zircon source testing methods from 1960 to 2021. Measured hydrological data are analyzed using statistical test methods, with the Mann–Kendall test employed to evaluate hydrological trends and the accumulation method utilized to identify abrupt change points in water and sediment data. The findings indicated a significant decreasing trend in annual runoff and sediment transport in the Huangfuchuan and Kuye River basins over recent decades. Abrupt change points in the runoff and sediment transport of the Huangfu River occurred in 1978 and 1998, showing a significant decline. In the Kuye River, abrupt change points in runoff occurred in 1979, 1997, and 2011. The first two points showed significant decreases, while the last point indicated a significant increase. Abrupt change points in sediment transport for the Kuye River appeared in 1979 and 1997, showing significant decreases. Detrital zircon experiments conducted on modern river sediments and river terrace sediments from several thousand years ago in the Huangfu River and Kuye River (2020) reveal that sediments are primarily composed of loess and bedrock, with modern river sediments containing a higher contribution rate of loess than terrace sediments.Results and Discussions The findings showed that precipitation and human activities are the primary factors influencing runoff and sediment transport. However, precipitation has not changed significantly in recent decades. Since 1980, human activities have become the dominant factor contributing to reductions in water and sediment, while precipitation plays a secondary role. The sharp decrease in sediment transport and the relative increase in the proportion of loess result from accelerated loess erosion caused by human activities. Simultaneously, exploiting surface water resources and soil and water conservation measures in the basin reduces runoff, decreasing the sediment transport ratio. Large amounts of slope-eroded loess are deposited in river channels rather than transported to drainage outlets, providing a sediment source for extreme rainstorms. Therefore, ongoing attention to soil and water conservation on the Loess Plateau is essential.Conclusions This study employs modern-scale hydrological observation data and geological sedimentology methods, combined with multi-scale approaches, to provide insights into controlling water and sediment disasters in the Yellow River. Although the application of detrital zircon source methods to reflect sediment transport in rivers remains underexplored, their effectiveness and applicability require further investigation. This study highlights the temporal changes in sediment transport in the Huangfuchuan and Kuye Rivers, key tributaries in the coarse and sandy area of the Yellow River. It proposes further analysis of spatial changes using InSAR, UAV, and other remote sensing technologies.关键词：Yellow River;runoff;Sediment transport capacity;Detrital zircon source198|175|0更新时间：2025-01-17
- “在河道植被区悬沙沉积研究领域，专家提出了基于随机位移方法的计算模型，为预测河道植被区悬沙沉积沿程分布提供解决方案。”摘要：In natural rivers, aquatic vegetation affects flow characteristics and suspended sediment transport patterns, resulting in a complex distribution of sediment deposition inside canopies compared to bare channels. The distribution of in-canopy deposition is challenging to simulate. This study examines the distribution of suspended sediment deposition in vegetated regions within open channels. The improved random displacement method is introduced to model water and sediment transport under the influence of emergent vegetation. In addition, a model is proposed to predict the distribution of suspended sediment deposition, specifically applicable within emergent vegetation canopies with real plant morphology (reeds). To verify the numerical model, sediment sizes, mean channel velocities, and plant densities are analyzed to evaluate the distributions of net deposition inside canopies with realistic morphology. Results indicated that the distribution of sediment deposition inside canopies is influenced by particle size, upstream mean flow velocity, and vegetation density. Particularly, the diminished deposition occurs at the canopy’s leading edge. The deposition patterns exhibit two distinct behaviors under varying sediment supply conditions in vegetated areas. The comparison between the predicted and measured deposition distributions demonstrates that the proposed numerical model provides an accurate simulation of deposition distribution when the upstream sediment supply is not limited to vegetated regions. The magnitude and trend of the model simulations align with experimental results. However, the model does not precisely calculate the distribution of sediment deposition when the upstream sediment supply is limited within vegetated regions. In conclusion, the proposed random displacement method-based numerical model accurately predicts the distribution of sediment deposition inside vegetated regions in open channels, provided the upstream sediment supply is not limited.关键词：vegetated channel;random displacement model;simulated reeds;sediment deposition distribution219|253|0更新时间：2025-01-17
SCIENCE OF MOUNTAIN RIVER CHANGE
- “最新研究进展显示，通过Logistic混沌映射与差分进化改进的人工蜂群优化算法，有效提升了水下节点定位精度，为水下定位技术发展提供了新方向。”摘要：Objective Localization algorithms for underwater acoustic sensor networks typically depend on distance estimation methods, such as point-to-point distance estimation or angle estimation, to achieve node localization between nodes. However, techniques like least squares may yield multiple coordinate values, leading to significant inaccuracies in node localization. To address this issue, it is essential to implement an optimization algorithm that can provide the optimal result, thereby enhancing the accuracy of node localization.Methods The artificial bee colony algorithm represents a category of intelligent optimization algorithm that imitates the honey harvesting mechanism observed in honey bees. It is relatively straightforward to implement and comprises four phases: population initialization, leading bee, following bee, and detecting bee. The artificial bee colony algorithm offers a novel approach to improving the accuracy of underwater node localization. This is achieved by transforming the optimization problem of node localization results into an optimization problem for the node objective function. However, the traditional artificial bee colony algorithm exhibited significant deficiencies, namely a slow convergence speed and a tendency to converge on local optima during the iterative process. To address these issues, the improved artificial bee colony optimization underwater localization algorithm by logistic chaos mapping and differential evolution (LDIABC), was proposed. This new algorithmic approach was developed with a particular focus on the population initialization phase and the leading bee search phase. Firstly, in order to address the issue of the traditional artificial bee colony algorithm being susceptible to falling into local optima due to excessive concentration or dispersion in the initialization of the population distribution, it was proposed to introduce logistic chaotic mapping into the population initialization phase of the proposed algorithm. The chaotic sequence generated by the mapping function was employed as a replacement for the random number generator, with the objective of achieving a more uniform distribution of the population during the initial distribution phase. Moreover, the discrepancy between the chaotic sequences generated by logistic chaotic mapping was demonstrated theoretically. Consequently, the proposed algorithm could avoid falling into local optima during the iterative process due to an overly concentrated population distribution, thus enhancing the global search ability of the proposed algorithm. Secondly, a fitness variance criterion was proposed to verify that the proposed algorithm was capable of avoiding the local optima problem during the iterative process. This further demonstrated the effectiveness of introducing the logistic chaotic mapping strategy into the population initialization phase. Subsequently, in the leading bee phase, a weight factor was incorporated as a parameter based on the variation strategy of the differential evolution, thereby improving the lead bee neighborhood search mode in the initial and final stages of the leading bee. By modifying the search mode in the initial and final stages of the leading bee, the leading bee was directed toward the optimal solution, thereby enhancing the global search efficiency of the leading bee and accelerating the convergence of the proposed algorithm. The theoretical implications of adjusting the weight factor for the leading bee during the initial and final stages of the optimization process were thoroughly examined. Finally, simulation experiments of the proposed algorithm and the comparison algorithm were carried out using the MATLAB software, employing identical simulation settings and datasets to evaluate the performance of the proposed LDIABC algorithm against its counterparts.Results and Discussions It was demonstrated that the LDIABC algorithm significantly improved the convergence speed and avoid falling into the local optima during the iterative process, thereby substantially enhancing the accuracy of node localization. In terms of convergence speed, the algorithm achieved convergence with 300 iterations, whereas the competing algorithm requires superior approximately 400 iterations to stabilize. Consequently, the LDIABC algorithm demonstrated superior convergence rates compared to the Tent-IABC, ELOABC, CODEGWO and SAPSO algorithms. Once the algorithms had reached a state of stability, the average localization error of the LDIABC algorithm was also the lowest, which indicated that the LDIABC algorithm effectively solved the issues present in the traditional artificial bee colony algorithm. In terms of the success rate of node localization, the LDIABC algorithm exhibits a higher success rate than the Tent-IABC, ELOABC, CODEGWO and SAPSO algorithms, with a success rate of up to 95%. In comparison, the four aforementioned algorithms demonstrated a success rate of between 80% and 86%. This indicated that the LDIABC algorithm increased the number of nodes with improved localization accuracy. In terms of optimized localization accuracy, the LDIABC algorithm demonstrated enhanced performance in comparison to the Tent-IABC, ELOABC, CODEGWO and SAPSO algorithms, with improvements of 6.36%, 13.33%, 14.16% and 16.88%, respectively. Moreover, in terms of maximum and average error, the LDIABC algorithm exhibited a lower localization error than the other algorithms, indicating an effective optimization effect.Conclusions The LDIABC algorithm has proven effective in improving convergence speed and preventing the process from becoming trapped in local optima during iterations. Consequently, this algorithm significantly enhances the accuracy of underwater node localization, resulting in impressvie optimization outcomes. By improving and enhancing the artificial bee colony algorithm and other intelligent optimization algorithms, it is possible to further optimize the results of node localization, thereby increasing the accuracy of node positioning and enhancing the value of the collected data. Furthermore, this development has advanced other intelligent optimization algorithms, allowing for deeper integration and application in optimizing parameters and model structures within underwater localization systems.关键词：artificial bee colony optimization;underwater localization;logistic chaotic mapping;fitness variance;weight factor222|210|1更新时间：2025-01-17
- “深度学习在基于WiFi的人体动作识别领域取得显著成果，提出了一种基于双重注意力机制和多通道、多尺度的时间卷积网络的动作识别方法，有效提高了动作识别的精度。”摘要：Objective With the continuous development of science and technology, cutting-edge advancements such as artificial intelligence and deep learning increasingly penetrate various fields, significantly improving social productivity. Among these, WiFi-based human action recognition has emerged as a prominent research direction, demonstrating essential application potential in smart homes, health care, military training, and other fields. However, with the diversified development of wireless communication technology, human action recognition faces new challenges, particularly in the expanding applications of multiple-input multiple-output (MIMO) systems. This necessitates in-depth research and innovation to ensure that human action recognition technology adapts to the diverse communication environments of the future. A MIMO system’s multivariate spatial diversity characteristics provide higher data rates and improved signal quality due to its design for parallel transmission through multiple channels. However, in practical applications, multi-channel parallel transmission often encounters interference from the multipath effect, causing the signal arriving at the receiving antenna to exhibit complex fluctuation characteristics with varying path lengths and incident angles. Since this path information is embedded in channel state information (CSI), the characteristics of CSI differ for the same action, while different actions may exhibit remarkably similar CSI characteristics. This results in incompleteness and generalization issues in feature extraction and action classification processes. Therefore, designing an effective mechanism to extract and classify human actions in complex MIMO environments is critical. This mechanism must overcome the multipath effect in multi-channel transmission to ensure accuracy and consistency when extracting action features. In this challenging context, innovative algorithms and model designs are crucial for addressing differences in CSI features between various actions and enhancing the model’s generalization ability and robustness.Method This study explores human actions’ physical and MIMO transmission characteristics, proposing a deep learning-based human action recognition method that employs a dual attention mechanism and a multi-channel, multi-scale fusion temporal convolution network to address the above challenges. Initially, a multi-channel information extraction model is constructed to leverage the spatial diversity inherent in MIMO systems. This model enhances the representation of data received for specific actions across different channels by extracting action-specific characteristic information from each antenna’s received signals. Simultaneously, a multi-scale integration mechanism is applied to fuse action characteristics at varying scales, bolstering the system’s ability to represent actions effectively. The extracted action features are then aggregated through a dual attention mechanism. The feature map fusion attention mechanism mines the correlation between action features from each channel, assigning higher weights to more relevant features and thereby enhancing the discriminative power of the extracted action representation. A temporal convolution network captures temporal dependencies within the extracted action features, distinguishing between actions with similar spatial characteristics but distinct temporal patterns. Compared to the feature map fusion attention mechanism, the feature channel attention mechanism assigns weights to different channels based on their importance for action recognition. This design allows the model to prioritize features significantly contributing to action recognition, enhancing its overall recognition capabilities. A temporal convolutional network processes the extracted action features. This network performs convolution operations on features at different instances, enabling the model to capture changes over time and identify long-term dependencies between action features. This capability is crucial for accurately recognizing complex and continuous actions. A global average pooling (GAP) layer is implemented to bridge the gap between the extracted feature maps and the action classifier. This operation preserves the action characteristics of each channel while facilitating a global comparison of these characteristics. Balancing the characteristics of each channel further improves the accuracy of action recognition.Results and Discussion Comprehensive experiments are conducted on public datasets and in real-world environments to evaluate the effectiveness of the proposed model. These experiments assess the model’s performance under controlled and uncontrolled conditions, ensuring its robustness and adaptability to practical scenarios. In the public dataset evaluation, the proposed action recognition model achieves an exceptional accuracy of 98.72% in identifying seven distinct human actions, surpassing the recognition performance of traditional models. This result highlights the model’s effectiveness in distinguishing different actions in controlled settings. Tests are also conducted in various natural settings to validate the model’s adaptability to real-world environments, including self-built laboratories, classrooms, and corridors. These environments present challenges such as uncontrolled lighting conditions, background noise, and varying distances between the user and the WiFi receiver. Despite these challenges, the proposed model maintains high performance, achieving accuracy rates of 97.94%, 97.28%, and 95.66%, respectively, for ten different actions in these real-world environments. These results demonstrate the model’s robustness and adaptability to real-world scenarios, making it a promising tool for practical applications. WiFi-based human action recognition offers significant potential for transforming domains such as healthcare, smart homes, and human-computer interaction. In healthcare, real-time action recognition enhances patient care, detects potential falls, and assists elderly individuals. Smart homes can evolve into intelligent living spaces, automatically adjusting conditions based on occupant activities. Human-computer interaction can also be revolutionized, enabling smoother and more natural interactions with emerging technologies. The development of this model provides reliable and intelligent human action recognition solutions for practical applications, fostering the deep integration of technology and society. This integration promotes innovation and paves the way for a connected and intelligent future.Conclusion The novel human action recognition model proposed in this study, which employs a dual attention mechanism and a multi-channel, multi-scale temporal convolutional network, represents a significant breakthrough in addressing the limitations of human action recognition in wireless environments. The model achieves remarkable accuracy in diverse environments by effectively capturing human actions’ spatial and temporal characteristics from channel state information (CSI) data. This study holds substantial theoretical value and practical guiding significance, paving the way for future advancements in human action recognition research. In future studies, further optimization and refinement of the proposed model based on feedback from various natural environments will enhance its adaptability and generalizability, enabling seamless integration with human activities and revolutionizing multiple domains, including healthcare, smart homes, and human-computer interaction.关键词：action recognition;deep learning;channel state information;TCN;attention462|347|0更新时间：2025-01-17
- “在教育研究领域，同行互评的评分可靠性和偏见建模取得新进展。研究人员提出了RPG6和RPG7模型，有效提升了主观题作业真实分数的估计准确性。”摘要：With the proliferation of many MOOC platforms, grading open-ended assignments submitted by many students presents a significant challenge in educational research. Peer assessment, which requires students to act as peer graders and evaluate their peers’ submissions of assignments, is the mainstream solution to address this issue. Researchers have recently proposed various probabilistic graph models to evaluate peer graders’ grading reliability and bias, effectively improving the estimated actual scores of assignments based on peer grades. However, the existing probabilistic graph models consider only the impact of students’ scores on the current assignment regarding their grading reliability, failing to account for their scoring deviation, which directly measures their reliability. This limitation affects the performance of these models. Therefore, this study proposes two novel probabilistic graph models, RPG6 and RPG7, which incorporate the peer graders’ grading ability, quantified based on their score deviation within a small proportion of submissions being spot-checked by teachers. These models, constructed on the foundation of two existing probabilistic graph models, represent the grading reliability of peer graders as a variable dependent on their scoring deviation-aware grading ability rather than their scores for the current assignment. This approach enhances the estimation of the true scores of assignments. Real classroom experiments demonstrated that the proposed RPG6 and RPG7 models achieve greater accuracy in estimating the true scores of assignments in peer assessment activities. Specifically, the RMSE values of RPG6 and RPG7 are, on average, 11.75% lower than those of the state-of-the-art method.关键词：peer assessment;probabilistic graph model;true score estimation;scoring deviation;grading ability;spot-checking275|237|0更新时间：2025-01-17
- “在医疗质量领域，专家提出了基于CTFN异构数据融合方法，结合患者出院小结文本与横断面数据预测再入院风险，为医院风险管理提供解决方案。”摘要：Objective Unplanned readmissions are a critical indicator for hospital risk management and a key measure of medical quality. Predicting readmissions poses a significant challenge within healthcare systems. While numerous existing methods have demonstrated remarkable results, there is still considerable room for further improvement. For instance, many studies have focused on a single structured data set or have used basic concatenation techniques to integrate heterogeneous data resources. These approaches do not fully utilize the extensive data available in electronic medical records and are often ineffective at integrating heterogeneous data resources. To address these issues, it is proposed to use machine learning technology combined with patient clinical heterogeneous data to develop a predictive model for unplanned postoperative readmission. Methods This model will identify patients at high risk of readmission at an early stage, enabling timely intervention. Such timely intervention will reduce medical financial and public medical resource wastage and alleviate the medical burden on patients. The discharge summary provides a comprehensive account of the entire process, from admission to discharge, and is the most crucial written document for describing the circumstances of the hospitalization. The Patient cross-sectional data, which includes three main categories: patient basic information, data generated during the patient’s hospital stay, and the patient’s past medical records, is strongly correlated with the likelihood of patients being readmitted to the hospital. This correlation has been corroborated by numerous scholars in experimental studies. By integrating these key patient clinical data sets with data mining, machine learning and other analytical techniques, it is possible to construct a predictive model for unplanned postoperative readmission. This model effectively identifies patients at high risk of readmission. This paper addresses two principal areas of enquiry, namely: 1) Address the issues arising from the absence of data and uneven distribution. The paper’s main goal is to conduct an objective analysis of the patient’s discharge summaries and cross-sectional data, identifying their characteristics and deficiencies. To address data loss and heterogeneous data structures, various approaches have been utilized. For missing values in cross-sectional data, a post-sample classification filling technique is employed. This involves filling discrete features of positive patients with the median and continuous features with the mean of that feature, with the same method applied to negative patients. Additionally, the issue of uneven distribution in patient cross-sectional data is resolved using the category weighting method. For text data, BCEWithLogit Loss is employed to counteract uneven sample distribution. 2) Propose a method based on CTFN for heterogeneous data fusion. The method begins by using the RoBerta model and a CNN model to extract representation matrices from the patient’s discharge summary text and cross-sectional data, respectively. These matrices capture the essential features of the data in a format suitable for fusion. The CTFN heterogeneous data fusion network is then used to fuse these representation matrices, obtaining comprehensive heterogeneous representation matrix. The CTFN is designed to build upon the TFN by directly calculating the tensor product to fuse the heterogeneous representation data. This method allows for the fusion of data without the need to expand the dimension of the unimodal representation matrix, thus preserving intra-modal information. To further enhance the fused representation, a CNN model is used to convolve the fused representation, amplifying important features. However, this process may result in the loss of intra-modal information due to the abandonment of single-modal representation dimension expansion before computing the tensor product of heterogeneous data representation matrices. To address this, a residual design is used. The fusion representation matrix is fused with the single-modality representation matrices of both discharge summary and cross-sectional data. These fused representation matrices are then concatenated to form the final heterogeneous data representation matrix. This approach compensates for the missing intra-modal information and enhances both intra-modal and inter-modal learning. The result is a more comprehensive expression of information in heterogeneous data, which ultimately improves the model’s performance.Results and Discussions The primary contribution is to propose the use of heterogeneous data in patients’ electronic medical records to comprehensively determine the likelihood of readmission, and to provide an effective methodology for the integration of heterogeneous data information. The CTFN method is capable of effectively integrating data from two different structures based on deep learning technology, thereby facilitating the acquisition of favorable prediction outcomes. The CTFN method was verified on clinical data from public hospitals, and the experimental results demonstrated the effectiveness of numerous improvements proposed in CTFN. Furthermore, this method exhibits the most optimal performance among all the compared models, with a markedly superior predictive efficacy. The recall rate is 76.1% and the AUC is 71.5%, both of which are superior to those of the baseline model under comparison. The method should satisfy the recall rate and accuracy requirement of real-world use, and provide technical support to assist doctors in determining the readmission goals of high-risk patients. The accuracy of using heterogeneous data to predict the risk of unplanned readmission of patients following surgery is significantly enhanced when compared to the use of a single structured data set.Conclusions This study proposes a predictive method based on CTFN for the fusion of heterogeneous data in the context of unplanned postoperative readmission. It has been demonstrated that in the context of predicting readmission for postoperative diseases, the incorporation of a greater number of highly relevant clinical data sets with varying structures and formats can effectively enhance the predictive effect of the model. Additionally, CTFN exhibits superior capabilities in heterogeneous data fusion compared with existing methods. The representation matrix of heterogeneous data generated through CTFN is more effective at representing contained within heterogeneous data, thereby enabling the model to make more accurate judgments based on a greater quantity of data. The risk of readmission is closely linked to a patient’s medical history, hospitalization recovery, and post-discharge social activities. The current experiment utilized a subset of medical records and focused on in-hospital recovery, which could be expanded to include more comprehensive data. Incorporating patients’ post-discharge social activities could further enhance the model’s predictive capability. Additionally, the model’s predictive performance could be improved by integrating medical imaging data, such as CT scans and X-rays, once they are properly fused with other data types using the CTFN approach.关键词：heterogeneous data;deep learning;tensor fusion;readmission;convolutional networks;residual structure384|350|0更新时间：2025-01-17
-
A Style Transfer Method for Chinese Landscape Painting Based on Detail Feature Extraction and Fusion AI导读
“在图像处理领域，研究者提出了SSTR网络，有效解决了中国山水画风格迁移中的图像质量低和细节特征丢失问题，为图像生成任务提供了新方案。”摘要：The goal of Chinese painting image style transfer is to render a real landscape scene image with Chinese painting artistic features, guided by a style reference, while maintaining the original, realistic scene image content. Recently, due to the rapid development of deep learning, convolutional neural networks (CNNs) and adversarial generative networks (GANs) have almost dominated image generation tasks, including style transfer. However, several uncontrollable problems persist, such as the loss of some semantics during the style transfer process, model collapse in GAN network training, and the checkerboard effect in CNN-based style transfer methods. The visual transformer model provides a new solution for image processing tasks, but its training requires a large amount of data and involves significant computational complexity. A Chinese landscape painting style transfer network, SSTR (swin style transfer transformer), is proposed based on the fusion of detailed feature extraction to address these issues and generate high-quality Chinese paintings. This approach introduces the Swin–Transformer within the StyTr2 network framework and uses the visual transformer to preserve the features of landscapes. In addition, the layered architecture of the Swin–Transformer and the sliding window attention mechanism are utilized to extract finer details of the artistic features of landscape paintings while reducing the model’s training complexity. Finally, a CNN decoder is incorporated after the Swin–Transformer decoder to refine the resulting image. The public visual dataset COCO and a public landscape painting dataset are employed for training, validation, and testing, with the results compared to several baseline methods. The experimental findings demonstrated that SSTR outperforms StyTr2 regarding style loss for the Chinese landscape painting style transfer task, showing superior feature extraction capabilities and image generation performance关键词：computer vision;Style Transfer;Visual Transformer;Swin–Transformer;Chinese painting;Twin-encoder327|242|0更新时间：2025-01-17
INTELLIGENCE INTERDISCIPLINARY SCIENCE AND ENGINEERING
- “在地质和工程领域，专家通过三维物理模型试验，揭示了库岸滑坡涌浪的传播规律，为滑坡涌浪灾害防治提供科学依据。”摘要：Objective Landslide surge disasters have consistently been one of the significant geological and engineering challenges studied both domestically and internationally. Accurately evaluating the process of water surges caused by scattered particle landslides on mountainous reservoir banks and understanding the propagation law of surges in rivers is crucial. This study analyzes the characteristics of surge propagation laws, establishes a formula for calculating the maximum height of surges, and verifies it through typical cases of water surges caused by scattered particle landslides, providing a scientific basis for disaster prevention and reduction.Methods This study focuses on the instability of reservoir bank landslides and the process of water surges, considering two scenarios: water landslides and submerged landslides. Three-dimensional physical model experiments of granular landslide surges are designed and conducted. The landslide surge simulation system, combined with the geometric shapes of river channels in the Three Gorges, Jinsha River Xiluodu, Yalong River Lianghekou Hydropower Station, and other reservoir areas, established a generalized river channel model of typical wide and straight reservoir sections in the reservoir area based on a geometric similarity ratio of 1:500. The model primarily comprises a source release device, a bank slope, and a closed glass water tank, which can simulate the entire process of landslide initiation, water inflow, accumulation, wave generation, propagation, and climb. Through physical model experiments, the process of landslide inflow and accumulation, as well as the characteristics of wave generation and propagation, are measured. Additionally, the amplitude and propagation characteristics of offshore waves propagating along the direction of landslide movement and lateral waves propagating upstream and downstream along the river channel are analyzed.Results and Discussions This study designed and conducted 96 sets of working conditions and 192 experiments to measure and analyze the process of landslide inflow movement and accumulation, as well as the characteristics of wave generation and propagation. The following conclusions are drawn: the entire process of granular sliding into water and wave propagation can be divided into three stages: the granular sliding into water stage, the wave generation stage, and the wave propagation and climbing stage. By analyzing the amplitude data of the surging waves, it was observed that the maximum peak and trough of the offshore surging waves were generated by the first wave. During the propagation of lateral waves, due to the collision and superposition of reflections on both sides of the river and solid waves, the maximum peak and trough of the surging waves appeared in subsequent wave trains. When a landslide transitions from a submerged condition to a waterborne condition, the maximum amplitude characteristic of the surge wave changes from a peak smaller than the valley to a peak larger than the valley. The wave amplitude generated along the direction of landslide movement is greater than that propagated along the reservoir bank, and the attenuation degree of wave amplitude height during lateral wave propagation is smaller than that of offshore waves. At the same time, this study compared and analyzed the inflow waveforms of two types of landslides: granular and block landslides. From the perspective of inflow forms, block landslides, and granular landslides exhibit significant differences, mainly reflected in aspects such as water tongue formation, impact pits, and wave amplitude. Compared to granular landslides, block landslides form more pronounced impact craters when entering the water. The upper water body projects toward the opposite bank in a tongue-like shape and splashes onto the water surface or bank slope, while the lower water body begins to move from rest. Block landslides, due to their interaction with water bodies, instantly enter the water with a larger volume, transferring more energy to the water body and generating larger wave amplitudes, which result in greater harm. A granular landslide forms a slender motion pattern on a slope, with a relatively small volume entering the water instantly, making it difficult to generate jet flow. Under block landslide conditions, the amplitude of the surge wave exhibits a characteristic where the peak value is significantly larger than the valley value. In contrast, under granular landslide conditions, the peak and valley amplitudes are approximately equal. This study analyzed the characteristics of the first wave and its maximum height during lateral surges. Under normal circumstances, when the surge propagates along the upstream and downstream of the river channel, the maximum height of the lateral wave is 2.5 times the height of its first wave. Based on the measurement system used in this experiment, 192 sets of data on the maximum height of swells generated by the inflow of test particles were obtained. The dimensional analysis method was employed to select three factors: the relative mass of the landslide body, the relative inflow height difference, and the relative submergence depth. The relationship between these parameters and the maximum height of swells was examined. The experimental results indicated that the maximum height of the surge is positively correlated with the mass of the granular landslide and negatively correlated with the initial submergence rate of the granular landslide. Within the dimensional range set in the experiment, it was found that the maximum height of the surge was negatively correlated with the height difference of the dispersed particles entering the water. The analysis revealed that compared to block landslides entering water as a whole, granular landslides exhibit a greater degree of discretization when entering water. During the unstable movement of landslides, strong collision and friction between granular particles result in partial energy loss. During the overall movement of the granular landslide, it continuously spreads to both sides. As the relative height difference of the inflow increases, the cross-sectional area of the landslide body impacting the water body gradually decreases under a single width. The reduction in kinetic energy transmitted by the single-width landslide body entering the water is more significant than the change in potential energy. Therefore, under indoor test conditions, the impact energy obtained by the unit water body is lower. For granular landslides without lateral constraints in the motion path, the maximum height of the surge is negatively correlated with the inflow height. The relationship between the maximum height of the surge and various influencing factors was obtained using dimensionless methods and multiple linear regression analysis. The empirical formula results showed good consistency with the physical experimental results. The empirical formula derived in this study was applied to the actual cases of the Hongyanzi landslide and the Chehalis Lake landslide surge disaster in Canada. The empirical formulas proposed by domestic and foreign scholars were compared and discussed. The results indicate that the empirical formula proposed in this study aligns well with other empirical formulas when compared to actual observation results, demonstrating a certain degree of reliability and applicability.Conclusions The experiment designed in this study primarily focuses on landslide surges in typical wide and straight sections of the reservoir area. The empirical formula proposed is applicable to granular landslides or loose accumulations and serves as a theoretical reference for landslide surge disaster warnings in reservoir areas.关键词：landslide surge;bulk body;physical model test;propagation characteristics;maximum height of surge;empirical equation220|333|0更新时间：2025-01-17
- “在滑坡涌浪灾害应急决策领域，专家提出了基于改进SPH的流固耦合分析模型，有效分析滑坡涌浪对大坝结构冲击响应规律，为大坝结构设计、施工及灾害应急防控提供重要参考。”摘要：Objective The SPH method is highly effective in handling large deformations and discontinuities in multiphase flow simulations, making it advantageous for modeling complex physical phenomena such as landslide-generated surges. It can more accurately simulate the interactions between landslide surges, terrain, and structures. However, in SPH–based fluid–structure coupling simulations, issues such as particle penetration often occur, particularly in cases involving large-deformation structures, where particle slippage may also present challenges. Consequently, existing studies frequently combine SPH with other techniques, such as the finite element method, to enhance performance. Additionally, the discrete element method cannot achieve fluid-structure coupling. Current research typically employs rigid bodies to simulate the landslide surge process, but no studies have simultaneously modeled the landslide body as a particle system while achieving a complete simulation of the coupling between landslide surges and large-deformation structures.Methods In the calculation process, the interaction between the fluid and the deformable body is initially performed by calculating the particles of the deformable body. Subsequently, the positions are corrected using spring forces. Based on this concept, the SPH method can be improved. In order to validate the accuracy of the proposed SPH method, a dam model test is conducted, utilizing a wave flume experimental setup to simulate surge waves. The experimental equipment measures 40.0 m in length, 0.3 m in width, and 0.3 m in height. The wave generator is driven by an AC servo motor capable of absorbing secondary reflected waves, allowing the simulation of small amplitude waves, solitary waves, and irregular waves. The dam is simulated using M15 micro-particle concrete, with steel reinforcement considered by reducing the modulus of elasticity. A compression test determined the elastic modulus of the dam model to be 70 MPa. The dam measures 20 cm in height, with a bottom width of 5.0 cm and a base width of 37.5 cm, and has a slope angle of 50°. Waterproof fabric is adhered to the model’s periphery and the flume to prevent water leakage around the dam structure.Results and Discussions The improved SPH method can reproduce the entire chain process of landslide collapse, wave generation, structural deformation, and the subsequent disaster sequence. This model, based on a spring-damping method, adjusts the density threshold of particles, effectively addressing penetration issues between solid and fluid particles at boundaries. It also enables the coupling simulation of fluid and structural large deformation. The calculation results of the improved method align well with indoor experimental results. The improved SPH method holds significant engineering application value in simulating the entire dynamic response sequence of landslide-induced waves on structures. Furthermore, the research findings provide technical guidance for emergency prevention and control of major projects affected by landslide-induced waves. By simulating water level changes when waves reach the dam structure, it is demonstrated that as the wave generated by the landslide propagates toward the dam, it swiftly surges, creating a shockwave that induces notable deformation in the dam structure. Subsequently, the wave undergoes fragmentation, leading to a reduction in the dam’s deformation under the effect of the fragmented wave. Afterward, the wave begins to act on the dam structure in the form of a long-period solitary wave. Therefore, in the stability design of dam structures, it is essential to consider the impact of different wave bands on the dam structure to ensure its resilience against potential threats. By analyzing factors such as the distance between the dam and the landslide and the angle of wave impact on the stability of the dam structure, it is found that when the distance between the landslide and the dam increases from 100 to 220 m, the maximum wave force on the dam decreases sharply by approximately 56%. As the distance between the landslide and the dam increases, the intensity of the wave and its force on the dam gradually diminish. Under wave impact, there is an increasing trend in wave pressure from the top to the bottom of the dam, and as the wave impact angle increases, the wave pressure on the dam structure gradually weakens. Significant displacement occurs at the top of the dam due to wave impact, which then gradually decreases. Therefore, in engineering design and disaster emergency processes, resilient measures can be implemented to mitigate the impact of landslide-induced waves, ensuring the safe operation of the dam.Conclusions This study proposes a fluid–solid coupling analysis model based on an improved Smoothed Particle Hydrodynamics (SPH) method. This model employs a spring-damping approach to manage fluid–solid coupling computations in expanding and contracting bodies, effectively resolving issues associated with complex fluid phenomena, large deformation coupling, and inter-particle penetration. Using this approach, the impact response laws of landslide-induced surges on the dam are analyzed.关键词：dam;SPH method;spring-damping model;landslide surge;impact response191|534|0更新时间：2025-01-17
- “在地质学领域，专家通过侧向渗透模型和数值模拟，揭示了花岗岩风化土内部侵蚀特性，为边坡失稳预测提供理论依据。”摘要：Objective The granite weathering process, influenced by internal and external forces from various geological effects, the distribution of irregular joints, and other factors, causes the granite weathering shell to exhibit differential weathering. This results in the formation of a granite binary structure of slopes, with the upper part of the slope consisting of weathered soil and the lower part of weathered rock and discontinuous differences between the upper and lower parts at the weathering interface. Weathered soil is the main source of granite binary-structured slopes, and clarification of its internal erosion characteristics is the basis for studying the damage mode and slip promotion mechanism of this type of slope. The internal erosion of granite-weathered soil under groundwater seepage involves fine particle migration and siltation. Due to the non-homogeneity of granite-weathered soil and the randomness of particle migration, existing studies cannot fully reveal the internal erosion process and particle migration law of granite-weathered soil. Additionally, there are fewer studies on the interaction process between fine particle transport and pore plugging inside the soil under dynamic head conditions, which limits further understanding of particle migration and internal erosion processes in granite-weathered soil.Methods This study focuses on the granite slope next to County Road X135 in Huangqi Town, Lianjiang County, Fujian Province. Understanding the geological phenomena of internal erosion in weathered soil clarifies the environment and triggering factors of internal erosion in granite-weathered soil. A lateral seepage test is conducted using a sandbox, combined with discrete element numerical simulations to reproduce the evolution of the internal erosion process. This reveals the dynamic processes of fine particle migration and siltation, investigates the influence of hydraulic conditions on internal erosion development, and summarizes the mechanisms of internal erosion in granite-weathered soil.Results and Discussions The results show that soil erosion is a gradual development process. In the initial stage of seepage, the bonding of fine particles at weak points in the soil is weak, making them highly susceptible to loss under water flow. Over time, the soil gradually stabilizes, and water flow preferentially follows established seepage channels. From a macroscopic perspective, more active water movement results in more developed seepage pore channels. At a finer scale, fine particle migration during erosion is irregular, expanding the pore structure while also causing blockages or deposition at the bedrock interface. Under dynamic head actions, fine particles continue to migrate, forming pore cavities within the soil, which eventually connect to form seepage channels, accelerating water transport. Increasing the amplitude of head change and the seepage inclination angle induces greater soil particle migration and loss, exacerbating internal erosion and prolonging the stabilization time of the soil. The composition and proportion of lost particles in weak zones are very similar to those of the upper weathered soil, indicating that the weak zones form through fine particle migration and deposition under seepage effects. This self-filtering process reflects the dynamic interaction of particle migration and deposition in granite weathered soil.Conclusions Under long-term rainfall, the fully weathered, fine-grained clay particles in granite weathered soil aggregate at different weathering interfaces. The long-term softening effect of groundwater forms weak structural surfaces that control slope stability. These weak structural surfaces divide the slope into an upper layer of granite weathered soil and a lower layer of strongly weathered granite, forming a binary structure. During the self-filtering process, the coarsening of soil particles caused by fine particle migration easily triggers soil collapse and damage. Additionally, the softening of the base-cover interface due to fine particle deposition evolves into a critical factor controlling slope stability. These factors are the primary causes of instability in granite binary-structured slopes. This study preliminarily clarifies the mechanisms of internal erosion in granite weathered soil and provides a theoretical basis for predicting slope instability and implementing disaster prevention and mitigation measures for granite slopes.关键词：model test;weathered granite;dynamic head;internal erosion165|317|0更新时间：2025-01-17
MECHANISM OF LANDSLIDE–DAMMED LAKE AND ITS CONTROL
- “最新研究进展表明，螺栓连接装配式墙板结构体系在低多层乡村住宅建筑领域具有显著优势，专家提出的抗震设计方法和建模技术，为该结构体系的工程建设和推广提供了理论基础。”摘要：Objective The bolt-connected precast concrete wall panel structural system provides multiple advantages, including speedy construction, convenient installation, trustworthy connections, disassembly, and repairability, which make it particularly suitable for low-rise rural residential buildings. In recent years, the engineering applications of this system have steadily increased. This research conducts studies on the bolt-connected precast concrete wall panel structure, including bolted connection node tests, wall panel tests, and structural system tests. The failure modes of the specimens are identified accordingly. The force transmission characteristics and seismic performance of the structure are systematically analyzed. In addition, calculation methods for the bearing capacities of nodes and wall panels are proposed. Numerical analysis software must be adopted to evaluate the integrity of the structural system and establish seismic design methods. However, existing solid element models or simplified calculation models based on the Opensees platform are challenging to converge during calculations, requiring a high level of professionalism from designers. Therefore, a simplified model of wall panels and bolted connection nodes is developed in this research to adapt to current engineering design software, facilitating the engineering application of this structural system.Methods This study simplifies a single bolted connection node into three translational degrees of freedom (x, y, z) nonlinear springs to simulate the mechanical behavior of the connection node and establish a simplified calculation model based on the analysis of the force transmission mechanism of bolted connection nodes. Formulas for calculating the tensile and shear bearing capacities of bolted connection nodes are proposed based on axial tensile and shear tests of bolted connection nodes. In addition, constitutive models for tension and shear of bolted connection nodes are established. Then, the effectiveness of the simplified model of bolted connection nodes is verified through quasi-static tests of bolted connection prefabricated wall panels. The simplified node model can be employed to investigate the seismic performance of the structural system. The current analysis and design software Midas Gen (2021 version) is utilized for system modeling. The specific modeling method involves using appropriate “plate elements” for prefabricated wall panels and floor slabs, “nonlinear spring” elements for bolted connection nodes, and setting compression springs only at the corner of the wall toe to simulate potential concrete corner crushing. Taking a demonstration project as an example, with a seismic fortification intensity of 7 degrees in the project area, a design basic seismic acceleration of 0.10g, and the third seismic group, the project employs a bolt-connected precast concrete wall panel structure. First, the modeling method is adopted to establish the design model of the structure. Elastic analysis under small and medium seismic actions and elastic-plastic time history analysis under large seismic actions are conducted to verify the inter-story drift angle of the structure. Elastic-plastic time history analysis is performed using El Centro, Kobe, and Artificially Generated Motion (AGM) waves, with the seismic response spectrum closely matching the design response spectrum specified in the code. The model can extract the internal forces of nodes under different conditions and calculate the bearing capacities of bolted connection nodes under the most unfavorable internal forces.Results and Discussions This study obtains force-displacement curves for the tensile and shear behavior of the bolted connection nodes in line with the conducted axial tensile and shear tests of bolted connection nodes. The mechanical behavior can be roughly divided into four stages: elastic, frictional slip, strengthening compression, and failure. Based on the constitutive models for the tensile and shear behavior of bolted connection nodes, a single bolted connection node is simplified into three translational degrees of freedom (x, y, z) nonlinear springs to simulate the mechanical behavior of the connection node and construct a simplified model of the bolted connection node. The simplified model of the bolted connection node is validated through quasi-static tests of bolted connection prefabricated wall panels. The validation results revealed that the peak bearing capacities simulated by the simplified model and the actual peak bearing capacity are 96.7 and 106.6 kN, respectively, with an error of –9.28%. Within a reasonable range of error, this indicates that the simplified model of bolted connection nodes can accurately simulate the bolted connection prefabricated concrete wall panel structure, predicting structural sliding deformation, peak bearing capacity, and lateral stiffness effectively. As illustrated by the calculations for the demonstration project, during the elastic analysis stage, the model’s natural vibration period is 0.08 s (in the Y direction), with high lateral stiffness. Under the effects of seismic intensities of 7°, 8°, and 8° fortification, the maximum inter-story drift of the structure is 1/9 999, 1/8 852, and 1/3 281, respectively, which are less than the elastic inter-story drift angle limit of 1/1 200 specified in the code. This demonstrates that the structure is in the elastic state. The results of the elastic-plastic time history analysis under major seismic events showed that the maximum inter-story drift angles of the design model are 1/1 446, 1/1 580, and 1/1 311, respectively, which are less than the required elastic-plastic inter-story drift angle limit of 1/120 specified in the code. These results indicate significant redundancy, meeting the seismic deformation requirements under rare seismic events. Taking an 8° (0.3g) fortification seismic event as an example, the verification of the bearing capacity of the bolted connection nodes showed that, at the most unfavorable location between the wall panel and the foundation, the maximum tensile and shear forces on the horizontal seam bolted connection node are 45.3 and 44.78 kN, respectively. These values are less than the design values of 260.37 kN for tensile and 115.24 kN for shear, demonstrating the safety and reliability of the bolted connection nodes.Conclusions Based on the axial tensile and shear tests of bolted connection nodes, this study proposes a simplified calculation model for bolted connection nodes, using three degrees of freedom nonlinear springs to characterize the mechanical behavior of the nodes. In addition, the constitutive models for the tensile and shear behavior of the node springs are proposed to describe the elastic-plastic performance of the springs. A comparison of this simplified model with quasi-static tests of bolted connection prefabricated wall panels exhibits desirable consistency, demonstrating the accuracy of the simplified model for bolted connection nodes. A modeling method for the bolt-connected precast concrete wall panel structure system is proposed. Appropriate “plate elements” are used for prefabricated wall panels and floor slabs, “nonlinear spring” elements are utilized for nodes, and compression springs are placed at the corners of the horizontal seam to simulate possible corner concrete crushing. Using a demonstration project as an instance, seismic design is systematically conducted, illustrating that the inter-story drift angle of the structure met the code requirements while the bearing capacity of the bolted connection nodes satisfied the seismic requirements. This demonstrates that the bolt-connected precast concrete wall panel structure system has good seismic performance.关键词：precast concrete wall panel structure;bolted connection;constitutive model;system modeling method;seismic design249|647|0更新时间：2025-01-17
- “在建筑领域，研究了现浇与套筒连接方式对陶粒混凝土复合墙板节点抗震性能的影响，发现套筒连接方式有效，整体工作性能良好。”摘要：Objective Developing prefabricated buildings effectively addresses the challenges associated with green buildings. A key issue in prefabricated buildings is the connection, as the quality of node connection performance directly affects the overall quality of these structures. Three cast-in-place components and one prefabricated sleeve connection member are designed to understand the influence of cast-in-place connections and prefabricated sleeve connections on the seismic performance of ceramsite concrete composite wallboard joints. All the components are L-shaped and primarily consist of a core area, a wall panel composed of ceramsite concrete slabs on both sides and a sandwich core insulation board. This component provides good thermal insulation performance, mitigating the issue of detaching external wall insulation boards.Methods For the proposed member, a low-cycle repeated load test is conducted, consisting of two parts: Vertical loading and transverse loading. The seismic properties of the components, including failure mode, hysteresis and skeleton curves, bearing capacity, stiffness degradation, energy dissipation capacity, and residual deformation, are investigated through these tests.Results and Discussions The test results indicated that the failure areas of both types of components occur on the flange plate and web. The concrete at the foot of the web tends to tear or crush, and the steel bars elongate or bend. Among the cast-in-place components, failure primarily results from crushing the concrete at the web’s base and the elongation and bending of the steel bars, ultimately leading to the specimen’s destruction. In contrast, the webfoot of the sleeve connection member shows less significant damage, with failure occurring due to the vertical reinforcement of the outer edge of the specimen’s web being pulled out. The core areas of the nodes in both types of components remain intact, meeting the design principle of “strong nodes and weak components”. The failure modes of both types of components are classified as bending and shear failure. The hysteretic curves of cast-in-place components and sleeve-connected components are similar. However, the hysteresis curves of sleeve-connected components exhibit greater “pinching” compared to those of cast-in-place components, primarily because the sleeve connection further limits the slip deformation of the components. A comparison of the skeleton curves of the two specimen types shows that the positive and negative yield loads of prefabricated sleeve connection members increase by 12.71% and 6.77%, respectively, while the positive and negative peak loads increase by 12.71% and 4.95%, respectively, compared to cast-in-place components. This indicates better bearing capacity performance for prefabricated sleeve connection members. Analyzing the displacement and load changes of the two specimen types during the “cracking to yield stage” (stage 1) and the “yield to peak stage” (stage 2) exhibits that the load value in stage 2 decreases compared to stage 1, while the displacement value increases significantly. This indicates that the relative deformation capacity of both component types is stronger in the later stages, enhancing their seismic resistance. Due to the brittle nature of the grouting aggregate, the ultimate deformation capacity of the sleeve connection member is affected to a certain extent, resulting in a slight decrease in the displacement ductility coefficient. However, based on the test results, the deformation capacity of the sleeve connection member is close to that of the cast-in-place component, which meets the deformation capacity requirements. The strong yield ratios of the cast-in-place specimens are comparable to those of the sleeve connection members, with the strong yield ratios of the two types of components being 1.39 and 1.36 in the positive direction and 1.95 and 1.92 in the negative direction, respectively. This demonstrates that the safety reserve capacity of the two types of components is the same. The stiffness degradation of the two types of components is also similar, with significant stiffness degradation occurring before yield and slower degradation after yield. The initial stiffness of the sleeve connection member is 7.53% higher than that of the cast-in-place component, and the overall stiffness curve is located above that of the cast-in-place member, indicating that the stiffness of the sleeve connection member is significantly higher than that of the cast-in-place component. The analysis results showed that the grouting sleeve plays a role in strengthening the lateral stiffness of the component, which is more conducive to the overall stress deformation of the specimen. The calculated cumulative energy dissipation and the equivalent viscous coefficient of both types of components increase with loading, representing the increasing cumulative damage of the components. The equivalent viscous coefficient of the grouting sleeve connecting member at each stage is smaller than that of the cast-in-place connecting member, indicating that the overall energy dissipation capacity of the sleeve connecting member is slightly inferior. The later failure of the cast-in-place component is primarily due to the crushing of the concrete at the foot, whereas the sleeve connecting member is constrained by the grouting sleeve, leading to reduced reinforcement slippage and smaller residual deformation after the peak load compared to the cast-in-place component. The strength degradation coefficient of the two types of components aligns closely with the strength degradation curve, and the strength change is relatively stable, indicating that the strength degradation capacity of the two types of components is essentially the same.Conclusions Therefore, the sleeve connection mode is effective, and the overall working performance of the components is satisfactory. Based on the advantages of low cost, ease of quality control, and effective shortening of the construction period, the prefabricated sleeve connection components can be a reference for applying such components in practical engineering.关键词：Low-cycle repeated test;Prefabricated type;seismic performance;wallboard nodes;Sleeve connection122|218|0更新时间：2025-01-17
- “在建筑节能领域，专家提出了预制混凝土夹心墙板轴心受压承载力计算公式，为提升墙体保温隔热性能提供解决方案。”摘要：The precast concrete panel combined with sandwich insulation technology significantly improves walls’ insulation and heat resistance while enhancing construction speed and energy-saving efficiency. This study designs a specimen of a bolt-steel plate connection precast concrete sandwich panel. It conducts an axial pressure load test to investigate its axial compression performance and proposes a calculation formula for the ultimate bearing capacity of the wall under the influence of the bolt-steel plate connection. The test results indicated that the failure mode of the sandwich wall panel is primarily the crushing of concrete above the bolt-steel plate joints and at both ends of the wall toe. Slight stratification between the concrete panel and the sandwich insulation layer is observed at the bottom of the wall, along with large vertical cracks on the side. The experimental axial load capacity of the specimen is compared to the formula proposed in this study and with existing formulas. The results showed that the differences between the test value and the calculated values of the existing formulas in China are approximately 5%, while the differences with the existing formulas in foreign countries ranged from 45.7% to 77.2%. The difference between the proposed formula and the test value is only 0.4%, indicating that existing foreign formulas are very conservative. The proposed formula accurately predicted the axial bearing capacity of the bolt-steel plate connection wall. Based on the experimental study, a finite element model is established to verify the proposed formula, considering parameters such as slenderness ratio, wall thickness, fitting arrangement, and concrete ribs. The analysis results demonstrated that the error between the numerical model and the proposed formula is within –8.05% to 2.65%, while the errors between the existing domestic and foreign formulas and the calculated values are within –2.25% to 8.58% and 41.61% to 78.41%, respectively. The prediction results of the proposed formula are more accurate. Finally, the influence of various parameters on the axial bearing capacity of the bolt-steel plate precast concrete sandwich wall is analyzed. The reinforcement ratio of the wall has no significant effect on the ultimate bearing capacity. A reasonable arrangement of GFRP connectors, vertical concrete ribs, a reduced slenderness ratio of wall panels, and an increased thickness of concrete panels enhance the bearing performance of walls.关键词：Precast concrete sandwich panel;bolt–steel plate connection;axial compressive performance;finite element analysis;bearing capacity calculation formula178|414|0更新时间：2025-01-17
PRECAST RURAL RESIDENCES
- “最新研究揭示液氮压裂技术对深部高温煤层气储层渗透性提升的影响机制，为煤层气开发提供理论参考。”摘要：The use of fracturing technology to enhance the permeability of coalbed methane reservoirs is an essential approach to strengthen coalbed methane development. For deep and high-temperature coalbed methane reservoirs, liquid nitrogen fracturing technology injects low-temperature liquid nitrogen into the coal seam, generating significant thermal stress. The “cold impact” effect effectively induces fracturing and increases permeability in coalbed methane reservoirs. This study examines the influence mechanism of liquid nitrogen treatment on the mechanical properties and damage characteristics of high-temperature coal-rock. Ultrasonic testing analyzes the changes in ultrasonic characteristics of high-temperature coal samples after liquid nitrogen cooling. Based on the Brazilian split test and acoustic emission monitoring, the influence of liquid nitrogen cooling on the physical and mechanical properties of high-temperature coal samples and the characteristics of acoustic emission are analyzed. In addition, the fractured characteristics of the damaged coal samples are quantitatively characterized based on the three-dimensional morphology scanning of the fracture surface and the apparent crack distribution to explore the influence mechanism of liquid nitrogen cooling on the failure mode of high-temperature coal samples. The experimental results indicated that the mechanical properties of high-temperature coal samples deteriorate significantly after liquid nitrogen cooling. The P-wave velocity decreases, and there is severe differentiation in ultrasonic frequency, with the primary frequency decreasing from 51.10 to 40.88 kHz. The coal samples gradually transform from brittle to ductile failure, with the highest decrease in the Brazilian split modulus and brittleness index reaching 67% and 75%, respectively. Under the Brazilian split load, the internal acoustic emission signals of the coal samples exhibit low RA values and high AF values, indicating that the internal failure mode of the coal samples is mainly tensile failure, followed by shear failure. With the increase in the initial temperature of the coal samples, multiple undulating regions form on the fracture surface of the coal samples, and the height difference between the maximum peak and the maximum valley in the three-dimensional morphology scanning reaches up to 37.60 mm. The liquid nitrogen cooling induces the cross-extension of microcracks, forming a complex crack network, which further leads to the formation of localized failure zones and the complexity of the fracture surface of the coal samples. The fractal dimension of the apparent crack distribution reaches its maximum value when the initial heating temperature is 100 °C, increasing from 0.50 at the initial temperature of 20 °C to 1.83. The research results provide a theoretical reference for the design of liquid nitrogen fracturing in deep and high-temperature coalbeds.关键词：liquid nitrogen;coal;thermal stress;Brazilian splitting;damage364|395|0更新时间：2025-01-17
- “最新研究揭示了煤矿地下水库煤柱坝体在浸水作用下的能量演化特性，为优化设计和安全控制提供了重要参考。”摘要：Objective Underground reservoirs in coal mines, with a wide prospect of application, are significant for water resource protection in coal mining in China. The stability of coal pillar dam bodies that are soaked in water for long or intermittent periods is central to the safe construction and stable operation of underground reservoirs in coal mines. Coal with various bedding angles is encountered during coal mining. Given this, the progressive catastrophic mechanical response of coal pillar dams in underground reservoirs under soaking is studied, and the influence of the bedding effect on the energy evolution characteristics of water-soaked coal under uniaxial compression is revealed.Methods The coal from the Shenmu Daliuta coal mine is selected as the experimental object, and X-ray diffraction results indicate that it is primarily composed of quartz, pyrite, and kaolinite. The bedding structure of the coal is apparent, and samples in different bedding directions are obtained by drilling along two directions perpendicular and parallel to the bedding surface. The initial water content state of the test sample before immersion is natural. The coal samples are immersed in static water to simulate the soaking of the coal pillar dam body. Pure water is used for immersion in the experiment to avoid the interference of special element composition with the test results. The experiment is divided into four groups based on the duration of immersion (0, 2, 4, and 16 d). The group with 0 d immersion is used as a reference, and each group contains one vertical bedding sample and one parallel bedding sample. The experiment is conducted using TEST 60, a rock multiaxial loading mechanical testing system at Sichuan University. The uniaxial compression test is performed using an axial deformation loading rate of 0.008 mm/min.Results and Discussions The energy evolution characteristics during the process of coal sample damage and failure are as follows: the damage process is divided into four stages: the initial damage stage, the stable damage stage, the abrupt damage stage, and the damage and failure stage. During the abrupt damage stage, due to the bedding structure of coal, the development of stress concentration between the layers of fractures leads to crack initiation and propagation, resulting in the sudden release of elastic strain energy accumulated in the coal samples. With the loading, the dissipative energy curve increases in a step-like manner, and correspondingly, the elastic strain energy curve decreases similarly. The energy evolution law of immersed bedding coal samples is as follows: vertical bedding and parallel bedding coal samples exhibit progressive failure characteristics during uniaxial compression loading, and the water immersion weakening process of parallel bedding coal samples progresses faster than that of vertical bedding coal samples. The energy index values of vertical bedding coal samples are generally higher than those of parallel bedding coal samples, consistent with the strength relationship. This indicates that the strength of vertical bedding coal samples is greater, and their ability to absorb strain energy is relatively stronger. The total energy corresponding to the peak stress of vertical bedding coal samples soaked in water for 0 days is about twice that of parallel bedding coal samples. After long-term immersion, the total energy and elastic energy of coal samples decrease significantly. The total energy corresponding to the peak stress of parallel bedding and vertical bedding coal samples immersed in water for 16 days is 56% and 53% of coal samples immersed in water for 0 days, respectively. The total energy of vertical bedding coal samples is higher than that of parallel bedding coal samples, and the rate of energy accumulation is faster. The energy evolution mechanism of immersed coal samples is as follows: the weakening of inter-particle cohesion, the dissolution of clay minerals, and the reduction of effective stress lead to a decrease in the strength of coal samples immersed in water for a long time and a reduction in the energy required for the expansion, aggregation, and penetration of internal cracks in coal samples. The failure surface of parallel bedding coal samples primarily extends along the bedding plane, making these samples more prone to failure and requiring less energy for crack propagation and penetration throughout the loading process. Under the same soaking time, the energy index values of parallel bedding coal samples are lower than those of vertical bedding coal samples. The failure surface of vertical bedding coal samples primarily extends along the axis, resulting in cracks that are mainly perpendicular to the bedding surface. The water absorption ability of the parallel bedding surface is stronger than that of the vertical bedding surface, jointly contributing to the differences in strength and energy evolution processes between vertical bedding and parallel bedding coal samples under short-term immersion.Conclusions After prolonged immersion, both vertical bedding and parallel bedding coal samples soften significantly, decreasing compressive strength and energy index values compared to non-immersion conditions. Therefore, the safety of coal pillar dam bodies subjected to long-term water immersion requires considerable attention. During the deformation and failure process of parallel bedding coal samples, the strength and energy index values are lower than those of vertical bedding under the same immersion duration. Greater emphasis should be placed on the safety of parallel bedding coal pillar dam bodies in underground reservoirs in coal mines. Due to differences in water absorption capacity and the relationship between failure and bedding surfaces in various bedding directions, the uniaxial compression energy evolution of different bedding coal samples varies after short-term immersion. The results enhance comprehension of the energy evolution behavior of soaked coal under load and provide a reference for ensuring the safe construction and stable operation of underground reservoirs in coal mines.关键词：soaking time;bedding;coal samples;uniaxial compression;energy evolution361|437|0更新时间：2025-01-17
DEEP UNDERGROUND SCIENCE AND ENGINEERING
- “在文本风格迁移领域，研究人员提出了基于流模型的新方法，有效改善了内容保留问题，并降低了内容与风格的不匹配度。”摘要：Objective In recent years, numerous style transfer methods in the text style transfer (TST) have relied on semi-supervised and unsupervised learning. The central concept of these methods involves mapping the text into a latent space, enabling the separation of content and style representations and facilitating style transfer. Text style transfer focuses on converting text attributes or styles while preserving the semantic content of the source sentences. Existing methods achieve significant progress in addressing the challenges associated with traditional TST tasks. However, these methods continue to face common limitations, including mismatches between the content and style of the transferred text and difficulties in maintaining the core semantics of the original text. Minimizing textual information loss during the transfer process remains critical to resolving content-style matching issues and preserving core semantics.Method This study introduces a flow-based text style transfer model. It adopts neural spline flows (NSF) as the foundational flow model for the text transfer task using the potential of flows in style transfer tasks. Flows employing affine coupling or autoregressive transformations provide accurate density assessment and sampling. The model comprises a conditional encoder, a series of invertible flows, a style discriminator, and two conditional decoders. The encoder (
$ X\to Z $ $ f:Z\to \hat{Z} $ $ {f}^{-1}:\hat{Z}\to Z $ $ Z\to X $ $ {x}_{\mathrm{a}}^{\mathrm{s}\mathrm{r}\mathrm{c}} $ $ {x}_{\mathrm{b}}^{\mathrm{s}\mathrm{r}\mathrm{c}} $ $ {\boldsymbol{z}}_{\mathrm{a}} $ $ {\boldsymbol{z}}_{\mathrm{b}} $ $ {x}_{\mathrm{a}}^{\mathrm{s}\mathrm{r}\mathrm{c}} $ $ {x}_{\mathrm{b}}^{\mathrm{s}\mathrm{r}\mathrm{c}} $ $ {\boldsymbol{z}}_{\mathrm{a}} $ $ {x}_{\mathrm{a}}^{\mathrm{s}\mathrm{r}\mathrm{c}} $ $ {\boldsymbol{z}}_{{\mathrm{b}}} $ $ {x}_{\mathrm{b}}^{\mathrm{s}\mathrm{r}\mathrm{c}} $ $ {{\textit{z}}}_{\mathrm{a}} $ $ {\boldsymbol{z}}_{\mathrm{b}} $ $ {\hat{\boldsymbol{z}}}_{\mathrm{a}} $ $ {\hat{\boldsymbol{z}}}_{\mathrm{b}} $ $ \hat{Z} $ $ {\hat{\boldsymbol{z}}}_{\mathrm{a}} $ $ {\hat{\boldsymbol{z}}}$ $ {\hat{\boldsymbol{z}}}$ $ {\boldsymbol{{\theta}} }=\mathrm{N}\mathrm{N}\left({\boldsymbol{z}}_{1:n-1}\right) $ $ {\hat{\boldsymbol{z}}}_{i}={f}_{{\theta }_{i}}({\boldsymbol{z}}_{i} $ $ {{\hat{\boldsymbol{z}}}_{\mathrm{b}{,1:n-1}}^{{*}}}={{\hat{\boldsymbol{z}}}_{\mathrm{a},1:n-1} }$ $ {\hat{\boldsymbol{z}}}_{\mathrm{b}}^{\mathrm{*}}=[{{\hat{\boldsymbol{z}}}_{\mathrm{a},1:n-1}},{{\hat{\boldsymbol{z}}}_{\mathrm{b},n:d}}] $ $ {\hat{\boldsymbol{z}}}_{\mathrm{a}} $ $ {\hat{\boldsymbol{z}}}_{\mathrm{b}}^{\mathrm{*}} $ $ {f}_{i} $ $ {f}_{i}^{-1} $ $ {\boldsymbol{z}}_{\mathrm{b}}^{\mathrm{*}} $ $ Z $ $ {\boldsymbol{z}}_{\mathrm{b}}^{\mathrm{*}} $ $ {\boldsymbol{z}}_{\mathrm{a}} $ $ {p}_{\mathrm{a}\to \mathrm{b}}\left(x|{\boldsymbol{z}}\right) $ $ {x}_{\mathrm{b}}^{\mathrm{r}\mathrm{e}\mathrm{c}} $ $ {\hat{\boldsymbol{z}}}_{\mathrm{b}} $ $ {\boldsymbol{z}}_{\mathrm{a}}^{\mathrm{*}} $ $ {p}_{\mathrm{b}\to \mathrm{a}}\left(x|{\boldsymbol{z}}\right) $ $ {x}_{\mathrm{a}}^{\mathrm{r}\mathrm{e}\mathrm{c}} $ 关键词：Text generation;Text style transfer;Neural network;Neural spline flows459|394|0更新时间：2025-01-17 -
Group Authentication Key Agreement Protocol Based on Certificateless Cryptography in UAV Network AI导读
“在物流行业，无人机技术应用广泛，但通信安全面临挑战。专家提出基于无证书密钥体系的群组认证密钥协商协议，为无人机通信安全提供解决方案。”摘要：ObjectiveUAV technology is increasingly applied in the logistics industry, the military field, and other scenarios. Due to the nature of certain tasks, the collected data must be exchanged between UAVs in scenarios requiring multiple UAVs to work together. However, the network communication environment is open and insecure, necessitating the use of a symmetric key to encrypt and decrypt data during the data exchange process. Ensuring that the UAV group negotiates the session key safely and efficiently becomes a key issue in the process. This study presents a group authentication key negotiation protocol based on a certificateless key system to meet the requirements of identity authentication and data privacy in UAV network communication. Diffie and Hellman proposed a key exchange algorithm that requires only one round of message exchange between the two sides of the session to calculate the same session key. Even if a malicious attacker eavesdrops on the data during the exchange, the attacker cannot compute the complete session key. However, this key exchange algorithm does not address the issue of a man-in-the-middle attack. If a malicious third party forges the identities of both session parties, it can determine two session keys with both parties, allowing it to arbitrarily steal and tamper with subsequently transmitted data. Due to the small size of UAVs, the computational and communication overheads they can bear must remain low, and the key agreement scheme must meet the requirements of lightweight scenarios. If directly applied to multi-party interaction scenarios, traditional two-party key negotiation protocols require numerous interaction rounds, resulting in significant computational and communication overhead. Therefore, group key agreement protocols for UAV networks must provide a lighter computing process and stronger security guarantees.Methods This protocol adopts a certificateless key system. As a third-party trusted server, the Key Generation Center (KGC) primarily facilitates identity registration before drone networking. Each drone member sends its identity ID to the KGC, which calculates the corresponding public-private key pairs based on the IDs in collaboration with the drone members. This protocol classifies the drone group into a central node with high computing power and multiple nodes with low computing power to reduce communication and computing expenses. The protocol comprises three stages: registration, negotiation, and joining/withdrawal. In the registration stage, a key pair is generated for each node. In the negotiation stage, the session key negotiation is conducted. The central node with high computing power in the group functions as the hub node, undertaking most computing and communication tasks. Once all nodes calculate their key pairs in the registration stage, only two rounds of interaction between the central high-computing-power node and the low-computing-power nodes are needed to complete the key negotiation. In addition, in scenarios where drones withdraw from or join the group, this protocol supports the dynamic alteration of the drone unit and group keys. Since the high-computing-power nodes bear most of the computational and communication overhead during the key negotiation process, they must be distinguished from the general low-computing-power nodes. The central and low-power nodes verify the digital signature to ensure identity reliability, and the protocol supports dynamic changes to the UAV unit and group keys. Once all nodes calculate their key pairs during the registration stage, only two rounds of interaction between the central high-computing-power node and the low-computing-power nodes are required to complete the key negotiation.Results and Discussions This study demonstrates the protocol’s security under the eCK model, which withstands guessing attacks, key duplication attacks, and forgery attacks in the eCK model. In the analysis, multiple query capabilities are granted to the adversary, and the proof concludes that if the CDH problem is difficult to solve, the adversary’s attack advantage is negligible. Based on the eCK model, this study establishes that the proposed protocol satisfies unforgeability, forward and backward confidentiality, and resists public key replacement attacks. A simulation experiment of the protocol is conducted to compare the calculation and communication costs with similar protocols. The simulation results indicate that the protocol achieves lower calculation and communication costs. Since the protocol does not rely on bilinear pairing construction, it is more suitable for lightweight UAV communication scenarios. In addition, this study evaluates the protocol’s performance against existing ones in terms of computational overhead, communication cost overhead, and security properties. A simulation verifies the actual performance of the protocol. The scheme employs secure hash function operations and elliptic curve scalar dot product operations with lower computational overhead, making it highly suitable for lightweight scenario requirements. Experimental results showed that both high-computing power nodes and low-computing power nodes in the scheme exhibit the lowest computational overhead. Whether in theoretical analysis or experimental simulation, the scheme shows good lightweight characteristics and high security, making it ideal for key negotiation scenarios of UAV swarms.Conclusions The proposed scheme divides the UAV cluster into a high-computing node and multiple low-computing nodes. The high-computing node undertakes most of the computational and communication tasks for the cluster, reducing the overhead of the other nodes and enabling the entire UAV cluster to negotiate the session key efficiently and quickly. This study also conducts a security analysis and proof under the eCK model, demonstrating that the scheme satisfies multiple security features, ensuring secure data transmission between UAVs. Finally, the analysis and experimental simulation of computational overhead and communication overhead confirm that the scheme exhibits good lightweight characteristics, meeting the requirements of UAVs with small sizes and low computing power.关键词：Unmanned aerial vehicles (UAV);Certificateless public key cryptography;identity authentication;Key agreement;Elliptic curve cryptography439|194|0更新时间：2025-01-17
CYBERSPACE SECURITY
- “在环境治理领域，研究人员开发了一种新型Co3S4/PES催化膜反应器，有效提高了过硫酸氢钾活化效率和活性氧物种利用率，为有机污染物降解提供了高效解决方案。”摘要：Peroxymonosulfate (PMS)-based heterogeneous advanced oxidation processes are restricted in practice by the low activation efficiency of PMS and the utilization rate of reactive oxygen species (ROS). A Co3S4/PES catalytic membrane reactor (CMR) is constructed using a flowing synthesis approach, employing a metal-organic frameworks (MOFs) template exchange strategy with ZIF-67 as the precursor and a polyethersulfone (PES) porous membrane as the substrate. The catalytic membrane is characterized by field-emission scanning electron microscopy (FESEM) and X-ray diffraction (XRD). The results indicated that Co3S4 nanoparticles with an average particle size of 23 nm are synthetically immobilized in situ within the pores of the PES membrane and are uniformly distributed along the membrane thickness, achieving an immobilization amount of 15.9%. The catalytic performance and reaction mechanism of the catalytic membrane are investigated by activating PMS to degrade a typical aromatic organic compound, Rhodamine B (RhB), as a model pollutant. Experimental results demonstrate that the Co3S4/PES catalytic membrane exhibits a RhB degradation rate exceeding 90% under conditions of a RhB solution concentration of 20 mg·L–1, an initial solution pH of 7, a temperature of 25 °C, and a membrane flux of 0.80 mL·min–1·cm–2 (corresponding to a residence time of 0.68 s). The apparent reaction rate constant is 97.83 min–1, and the turnover frequency (TOF) reaches 489.15 L·min–1·g–1, both of which are 2 orders of magnitude higher than those achieved using the conventional suspension batch treatment mode with Co3S4 powder. The Co3S4/PES catalytic membrane also maintains good stability. The dispersion of membrane pores effectively prevents the aggregation of Co3S4 nanoparticles while confining the reaction within the micro-nano-scale membrane pores under flow-through conditions. This enhances mass transfer and contact between reactants and catalysts, accelerates PMS activation, and facilitates the efficient generation of singlet oxygen (1O2), which plays a dominant role in the rapid degradation of RhB.关键词：Catalytic membrane reactor;Advanced oxidation processes;Rhodamine B;Peroxymonosulfate;Flowing reaction;ZIF-67;Co3S4127|54|0更新时间：2025-01-17
- “最新研究发现，电动导排孔隙水能有效修复氮磷污染沉积物，具有应用前景。”摘要：Pore water plays a key role in sediment nutrient exchange. This study uses electrokinetic geosynthetics (EKGs) as electrodes to drain sediment pore water in the presence of overlying water and conducts five sets of indoor experiments (one control group and four groups exposed to voltage gradients of 0, 0.5 (intermittent power), 0.5 (continuous power), and 1.0 V/cm (intermittent power)) to investigate the performance and mechanisms of pore water drainage in regulating sediment nutrient release and the response of microbial communities. The results showed that draining pore water significantly inhibits the release of nitrogen and phosphorus at the sediment-water interface (SWI). However, the responses of nitrogen and phosphorus exhibit differences. The TN release flux at the SWI shows a gradual downward trend, ranging from –15.698 to 79.903 mg/(m2·d). Within 14 days of the experiment, the TN release flux reduces to within 10% of the maximum release flux. Compared to the control, conducting pore water drainage alone reduces the nitrogen release flux at the SWI by 76.72%. In contrast, the TP release flux at the SWI fluctuates within the 28-day experiment, ranging from –3.558 to 4.279 mg/(m2·d). After the experiment, the pH and LOI of the sediment do not change significantly. TP decreases by approximately 1.06%～5.02% following drainage, while the reduction in TN content occurs primarily near the anode, decreasing by 2.65%～13.63% compared to the initial state. The drainage causes a nonsignificant decline in community abundance indices (the Chao1 index and ACE index of microorganisms in the sediment are slightly reduced) but does not change the dominant species of sediment microorganisms. The dominant phyla are Proteobacteria (13.41%～24.40%), Bacteroides (12.77%～29.71%), and Chloroflexi (5.01%～18.64%). The pH and water content are the primary factors affecting microbial community structure, while the LOI and conductivity of the sediment are secondary factors. The total phosphorus content in the sediment has the least impact. The research demonstrates that the electric drainage of pore water is an economical and low-ecological-impact technology for restricting in-situ sediment nutrient release.关键词：pore water;sediment;electrokinetic remediation;release flux;microbial biodiversity81|102|1更新时间：2025-01-17
- “在环境治理领域，专家提出了细菌氧化—pH分段沉淀—循环沉淀处理酸性矿井涌水新技术，为解决环境问题提供解决方案。”摘要：Objective Acid mine drainage (AMD) has strong acidity and high heavy metal content, which poses a significant threat to the ecosystem and human survival. The proper treatment of AMD remains one of the major environmental problems requiring urgent resolution. The biological method has gradually become a research hotspot due to its advantages of low cost and environmental friendliness. However, research on the practical application and resource utilization of biological AMD treatment remains in the small-scale exploration stage. This study proposes a new integrated technology to address the problems of excessive heavy metal ions in acidic wastewater, high sludge disposal costs, and limited comprehensive treatment technologies.Methods A new technology involving bacterial oxidation-pH fractional precipitation-cyclic precipitation for AMD treatment is proposed, and semi-industrial application is conducted. Using this method, a pilot plant is set up at a pollution source site in Guizhou Province to investigate the effects of temperature, aeration intensity, pH value, microbial community structure, and influent Fe2+ concentration on AMD treatment efficiency. In addition, the effects of different precipitating agents, precipitating pH value, and cycle times on the grade of iron in the recovered slag are examined.Results and Discussions The research results show that iron and aluminum are the main pollutants of acid mine drainage. The treatment efficiency of the bioreactor is positively correlated with temperature and aeration and negatively correlated with initial Fe2+ concentration. Temperature is a principal factor affecting the efficiency of bioreactors. Although the maximum flow rate remains 0.33 m3·L–1 at a low temperature of 10 ℃ and the oxidation time is 13 h, the oxidation efficiency more than doubles when the temperature rises to 25 ℃. If the discharged wastewater meets the standard, a higher flow rate corresponds to a shorter residence time of the wastewater in the reactor, which means a higher treatment efficiency of the bioreactor. At a low temperature of about 12 ℃, the bioreactor maintains a good oxidation effect and a significant flow rate for Fe2+: 100～1 000 mg ·L–1 AMD, with a Fe²⁺ removal rate as high as 410 g·h–1. The pH value significantly influences the removal of iron and aluminum in AMD. Fe3+ is removed when pH = 3.5, Al3+ is released in the range of pH = 4～5, and all indicators meet the emission standard when the solution pH = 5. The comparison test of different termination pH values finds that alkali consumption does not correspond to changes in pH values because the metal ions in the wastewater after oxidation are mainly Fe3+ with a small amount of Al3+. When the pH value of wastewater is less than 4, OH– in the alkaline reagent combines particularly with Fe3+, while the pH value is primarily connected with Al3+ in the wastewater. The remaining OH– only acts to increase the pH value, so the alkaline reagent is mainly consumed at low pH. The treatment efficiency of the bioreactor increases with the aeration rate. When the aeration rate increases from 16 to 32 Nm³·h–1, the rise in bioreactor oxidation efficiency is significantly less than when the aeration rate increases from 8 to 16 Nm³·h–1. Although high aeration intensity can improve the dissolved oxygen level, it also increases the impact force on the filter material, which can influence bacterial attachment to the filter material. This makes it complex for bacteria to attach to the filter material and can cause additional strain loss. Therefore, the aeration intensity cannot be increased blindly. Combining operation cost and treatment efficiency, the optimal reaction condition is 16 Nm³·h–1 aeration and pH = 5. Lime milk is the best alkaline reagent to regulate pH value, and the removal rates of Fe2+ and Al3+ in the pilot system are above 99%. In addition, microbial classification and sequencing results show that the relative abundance of Proteobacteria increases under heavy metal pollution and has adaptability to acidic environments. During the experiment, Proteobacteria maintain absolute superiority at the phylum level at all periods. The Acidophilus and Acidophilus in the original bacterial fluid are replaced by Unclassified_Callionella, Ferredoxin, and Acidophilus at the genus level. The iron-oxidizing bacteria replace Callionella as the final dominant bacterial group in the bioreactor, indicating that the iron-oxidizing bacteria have a competitive advantage and better adaptability to high iron ion concentrations. In the pilot process, the biological method effectively treats the acid mine wastewater and successfully separates the iron in the waste residue, which is conducive to subsequent resource utilization. With the increase in cyclic precipitations, the iron and aluminum content in the waste residue also gradually increases and decreases after reaching maximum values of 0.67% and 20.26% on the 8th cycle. The minimum particle size of the precipitate reaches 258 nm before increasing again, showing a trend of first decreasing and then rising. Excessive circulation can produce a “dissolution” phenomenon, causing some of the target substance to dissolve back into the supernatant or mix with impurities in the precipitation, thus reducing the purity of the target substance. Therefore, the number of cyclic precipitations should be controlled based on the actual situation, and the optimal purity of the filter cake can be achieved with about eight cycles of cyclic precipitation in this test.Conclusions Filter cake with a Fe content of 41.15% can be enriched through process optimization involving the combination of stage and cycle precipitation. It is expected to be used as raw material for applications such as fertilizer, cement, and flocculant, achieving economic benefits in construction, environmental protection, and other fields. This study confirms the feasibility of efficiently treating AMD with oxidizing bacteria as the core and discusses various pilot test parameters. The findings provide a more reliable basis and reference for the future large-scale industrial application of AMD biological treatment.关键词：Acid mine drainage;Biological oxidation;Resource utilization150|197|0更新时间：2025-01-17
ENVIRONMENTAL ENGINEERING
- “在近海工程建设领域，广州沿海工程专家利用纳米SiO2改良水泥土，有效提高其力学性能，为沿海工程提供参考。”摘要：Objective Most construction projects in offshore areas are affected by seawater corrosion and dynamic loads. Many of these projects have caused a series of accidents due to the corrosion of cement soil, such as the swelling and collapse of road bases and the cracking and destruction of foundations. From the perspective of dynamic load, cement soil is subjected to various stresses, including earthquakes, typhoons, vehicle-induced crushing, wave impacts, and other loads during construction and use, which inevitably aggravate the deterioration of the cement soil foundation.Methods In order to improve the mechanical properties of cement soil foundations in offshore areas, a coastal engineering clay from Guangzhou is selected to prepare nano–SiO2-improved cement soil specimens. The GDS true (dynamic) triaxial instrument is employed to conduct cyclic loading tests on the improved cement soil. The specimens are saturated in a vacuum using a saturator, placed in a seawater solution for corrosion after saturation, and then subjected to counterpressure saturation using the GDS true (dynamic) triaxial instrument. The peripheral pressure is set at 200 kPa, with a loading frequency of 1 Hz (sine wave), an initial axial force of 2 200 N, and a dynamic stress amplitude (σd) of 100, 200, 300, 400, 500, 600, 700, 800, 900, and 1 000 kPa. The specimens are loaded in ten stages and vibrated 10 times in each stage to investigate the effect of nano–SiO2 on the mechanical properties of cement soil under corrosion conditions. To study the dynamic elastic modulus and damping characteristics of the improved cement soil under varying corrosion durations and concentrations, the changes in the dynamic elastic modulus of cement soil before and after improvement are compared. A dynamic elastic modulus ratio decay model is established, and the influence of changes in the damping ratio is analyzed.Results and Discussions The results showed that under the combined effects of cyclic loading and seawater corrosion, the dynamic elastic modulus of cement soil before and after nano–SiO2 improvement decreased with increasing seawater concentration. When the dynamic strain (εd) was less than 0.3%, the rate of decrease of the dynamic elastic modulus was faster, whereas the decrease rate slowed when εd exceeded 0.3%. Under similar corrosive conditions, the dynamic elastic modulus of cement soil increased gradually with the rise in the corrosion time, but the growth rate decreased over time. The dynamic elastic modulus of nano–SiO2-improved cement soil was consistently greater than that of ordinary cement soil in all corrosion cycles. A good linear relationship was observed between the inverse of the dynamic elastic modulus and the dynamic strain of the cement soil when the Konder model was used. The maximum dynamic elastic modulus (Ed0) of nano–SiO2-modified cement soil was significantly higher than that of ordinary cement soil. The Darendeli model was utilized to analyze the dynamic elastic modulus of cement soil, and the decay model of the dynamic modulus ratio of cement soil before and after nano–SiO2 improvement was obtained. This model better describes the variation in the dynamic elastic modulus of cement soil with dynamic strain under a corrosive environment. The hysteresis loop area (S) increases continuously with the increase in dynamic strain. The S–εd curve gradually moves downward with time and rises with the increase in sea salt concentration. Under the corrosion of seawater solutions with different concentrations, the damping ratio (λ) of cement soil before and after nano–SiO2 modification increases with dynamic strain and gradually levels off, while the λ–εd curve shifts upward as the corrosion concentration increases. The overall trends of the S–εd and λ–εd curves for nano–SiO2-modified cement soil are slower than those for ordinary cement soil. For the same corrosion time, the hysteresis loop area (S) and damping ratio (λ) of nano–SiO2-modified cement soil are smaller than those of ordinary cement soil and increase with the increase in the concentration. After mixing an appropriate amount of nano–SiO2 into the soil, nano–SiO2 reacts with cement to generate cementitious substances that fill and refine some of the soil’s pores. This enhances the structural stability of the soil, increases the force between particles, improves load transfer efficiency, reduces energy loss during loading, and lowers energy consumption. A large quantity of cementitious substances is rapidly produced on the surface of soil particles. These substances, which appear as rods, bars, and spheres, increase with the concentration. Compared to ordinary hydraulic soil, the surface of soil particles rapidly produces a large number of gel-like substances in the form of rods, bars, and spheres that stack together to fill pores and particle gaps. This forms a dense spatial network, effectively enhancing the strength and structural stability of cement soil, reducing seawater erosion, improving corrosion resistance, and increasing resistance to the coupling effects of dynamic loading and seawater corrosion. Most current research on nano–SiO2 cement soil focuses on its engineering properties under static load in ordinary environments, while studies on its dynamic properties in seawater corrosive environments remain limited.Conclusions This experiment investigates the effects of different seawater concentrations and corrosion durations on the kinetic properties of nano–SiO2-modified cement soil under the coupling of dynamic loading and seawater corrosion. It reveals the deterioration patterns of the mechanical properties of cement soil under these conditions and examines the improvement mechanism of nano–SiO2, providing a reference for future studies on cement soil base structures in coastal areas.关键词：sea salt solution;cement soil;nano–SiO2;dynamic triaxial test;dynamic elastic modulus;damping ratio182|233|0更新时间：2025-01-17
- “在煤矿安全领域，专家分析了联络巷防水密闭墙稳定性，确定了最大安全水头高度，为矿井防水密闭墙稳定性评估与位置优化提供理论基础。”摘要：Objective The waterproof sealing wall in the connecting roadway is affected by two mining operations, and its stability is related to the safety of adjacent mining working faces.Methods Firstly, theoretical analysis was performed to calculate and analyze the maximum safe water head height the waterproof sealing wall can withstand without being affected by mining activities. In addition, the structural weak plane between the waterproof sealing wall and the coal rock column, which is prone to instability, was identified. By generalizing the physical model of waterproof sealing walls, the water pressure resistance of these walls was calculated in terms of compression resistance, shear resistance, and impermeability. Secondly, numerical simulation was used to analyze the stress state, displacement, and contact surface sliding of the waterproof sealing wall under one or two mining operations at different water head heights (6.0 m, 9.0 m, 12.0 m, and 15.0 m). Using numerical models, the stress state, deformation, and plastic failure zone distribution of the sealed wall were simulated and studied as the working face approached and passed the wall’s location. By assigning different water head heights to the goaf, the damage and stability of the sealed wall under mining and water pressure conditions were evaluated. Based on an analysis of structural weak planes, a self-developed stress and displacement monitoring system for the waterproof sealing wall in the underground goaf was employed. This system was used to strategically arrange and monitor measuring points on the sealing wall. The stress or displacement monitoring device was placed in direct contact with the waterproof sealing wall to record the stress and displacement of the wall. The data acquisition device included a memory unit and a collector. The memory was connected to the stress or displacement monitoring device to store the data, whereas the collector was electrically connected to the memory to retrieve the monitoring data. An analysis device was connected to the acquisition device to process the data and evaluate the stress or deformation state of the waterproof sealing wall. The maximum safe water head height of the waterproof sealing wall under mining and water pressure conditions was determined. In addition, a borehole stress meter was used to measure the distribution pattern of lateral stress in coal pillars, revealing the distribution law of lateral support pressure in mining face 3-1. This data was used to optimize the placement of the waterproof sealing wall.Results and Discussions Without considering the impact of mining, the maximum safe water head height of the waterproof sealing wall is 12.4 m. Structural weak surfaces, such as the top and bottom corners of the wall and the contact surface with the coal pillar, are prone to damage. Recent stress and displacement monitoring of the sealed wall indicated it is generally in a stable state. However, weak structural surfaces are commonly observed at the contact surface between the sealing wall and the coal pillar, as well as at the bottom corner of the sealing roof. Therefore, these areas require enhanced monitoring or observation. Through theoretical analysis, numerical simulation, and consideration of mining and water pressure, the maximum water head height for the sealed wall was determined to be 9.0 m.Conclusions A warning water level was set at 80% of this value, making the warning water head level for the 31313～31315 connecting roadway 7.2 m. Monitoring results confirm that the sealing wall is currently in a stable state. The peak area of lateral pressure on the working face lies between 10.5 and 12.5 m. To avoid peak support pressure, it is recommended that waterproof sealing walls be positioned more than 15～20 m away from the goaf side. These findings provide a theoretical basis and practical reference for the stability evaluation and placement optimization of similar waterproof sealing walls in mining operations.关键词：Mine water hazard;waterproof sealing wall in liaison lane;water pressure resistance;side abutment pressure;location optimization133|383|0更新时间：2025-01-17
- “在地下工程领域，专家通过模拟试验和数值模型，研究了天然破裂砂岩的冲击强度特性，为保障工程结构稳定提供重要参考。”摘要：Studying the strength characteristics of “Natural” fractured sandstone affected by impact direction is of great importance to ensuring the long-term stability of underground engineering structures. In this research, the “Natural” distribution of fractures is initially obtained through biaxial compression simulation, and the distribution characteristics of fractures under varying confining pressures are analyzed. Subsequently, the “Natural” fracture distribution is extracted and rotated by 0, 90°, 180°, and 270° as a whole for transplantation into the SHPB numerical model of the Particle Flow Code. On this basis, the relationship between the impact strength of “Natural” fractured sandstone and factors such as static compressive strength, fracture distribution information, and impact direction is examined through dynamic loading simulation. The results indicate the following: 1) When the biaxial compression confining pressure is low, the sliding surface of the specimen is relatively distinct. As the confining pressure increases, the sliding surface becomes more challenging to identify. 2) With the increase in the biaxial compression level and confining pressure, the impact strength and corresponding time of the specimen exhibit a decreasing trend. When the impact strength exceeds 20 MPa, the tangent slope of the stress–strain curve in the pre-peak elastic stage of the specimen remains consistent. 3) When the impact direction is perpendicular to the maximum principal stress direction of biaxial compression, the number of fractures has a more significant impact on the impact strength. In contrast, when the impact direction is parallel to the maximum principal stress direction of biaxial compression, the influence of the fracture number on the impact strength is relatively minor. 4) To further analyze the influence of the fracture number on impact strength, the concept of the inverse value of the distributed fracture number is introduced. Under different impact directions, the variations of the inverse value and the impact strength with the biaxial compressive strength of the specimen are generally synchronous. 5) The dependence of impact strength on the impact direction is closely associated with the macroscopic fracture surface. The presence of the macroscopic fracture surface effectively impedes the transmission of stress within the specimen, changing the propagation path and range of the impact stress under different impact directions.关键词：impact direction;“Naturally” distributed fracture;strength characteristic;numerical simulation of Particle Flow Code;fracture extraction, rotation and transplantation103|144|0更新时间：2025-01-17
- “在建筑结构领域，专家提出了基于各向异性混合核函数高斯过程回归的RC柱概率抗剪承载力预测模型，有效描述了抗剪承载力的不确定性，并校准了传统模型的预测精度。”摘要：A probabilistic shear strength model for reinforced concrete (RC) columns is proposed based on Gaussian process regression (GPR) with an anisotropic mixed kernel function to address the limitations of traditional models, which often exhibit low nonlinear approximation ability and poor generalization performance. A new anisotropic mixed kernel function is developed using the additivity and autocorrelation properties of the Matern and Rational Quadratic kernel functions, while the automatic relevance determination function is introduced to account for the effects of various feature parameters. The probabilistic shear strength model for RC columns is established by integrating the anisotropic mixed kernel function with the Gaussian process regression algorithm. The posterior distribution of model hyperparameters is obtained using Bayesian inference, and the hyperparameters of the probabilistic shear strength model are determined through the maximum likelihood estimation method. The effectiveness of the proposed model is validated by comparing it with traditional kernel functions, machine learning models, and mechanical models using 91 sets of experimental data. The analysis results indicated that the deterministic prediction indices RMSE and MAE of the proposed model are reduced by approximately 16% and 19%, respectively, compared to traditional kernel functions. In contrast, the probabilistic prediction indices NLPD and MSLL are reduced by about 15% and 23%, respectively. Compared to traditional machine learning models, the RMSE and MAE of the proposed model are decreased by 38% and 39%, respectively. The proposed model demonstrated high prediction accuracy and generalization performance and effectively quantified the uncertainty in the shear strength of RC columns.关键词：reinforced concrete columns;anisotropic compound kernel function;Gaussian process regression;probabilistic shear strength model;uncertainty178|232|1更新时间：2025-01-17
CIVIL ENGINEERING
-
Review of Demand Response-based Optimal Scheduling of Electric and Thermal Integrated Energy Systems AI导读
“综合能源系统灵活性调节能力下降，需求响应成为提升电网互动灵活性的重要途径。专家综述了基于需求响应的电热综合能源系统调度研究，为优化调度提供参考。”摘要：With the changes in energy demand and structure, the integrated energy system faces a marked decline in flexibility adjustment capacity while meeting customer demand and securing energy supply. Demand response serves as an essential approach for the demand side to engage in grid stability coordination, improving the flexibility of the integrated energy system and compensating for the lack of system flexibility through demand-side synergistic optimization of the coupled and complementary forms of multi-energy. This study provides an overview of the current research status, classification of scheduling models, and model solution methods for demand response-based scheduling of electric and thermal integrated energy systems in recent years. Firstly, the current research status of demand response mechanisms at the domestic and international levels is analyzed. Based on the different classification criteria of current demand response mechanisms, demand response mechanisms are classified into two categories: based on the guiding method and the evaluation method of the user's contribution to the system. They are divided into tariff-type and incentive-type demand responses based on the guiding method and tariff-type non-direct evaluation and baseline and quasi-linear demand responses in direct evaluation. In particular, compared to tariff-based demand response, which is greatly affected by electricity price, incentive-based demand response does not involve the setting of tariffs and is more capable of fully mobilizing many consumers to actively participate. However, due to the current single incentive method of incentive-based demand response, it cannot fully realize its significant regulation potential. Compared to baseline demand response, quasi-linear demand response more effectively raises positive interaction between the source and load sides and facilitates new energy consumption during multi-source coordinated scheduling in the context of large-scale multi-user participation. However, the impact of uncertainty factors such as wind power on load collinearity has not yet been addressed in-depth and requires further study. Secondly, the composition and primary characteristics of the electric-thermal integrated energy system are analyzed. The analysis reveals that the integrated electric and thermal energy system is a power system with close multi-energy coupling. Based on this, the current research status of the three kinds of integrated electric and thermal energy system scheduling models, including the basic, flexibility, and stochastic models, classified based on differences in application scenarios, is elaborated. A comparative analysis of the adaptive scenarios, advantages, and disadvantages of the current demand response-based optimal scheduling models for electric-thermal integrated energy systems is conducted. Current scheduling model solution methods are mainly classified into two types: analytical methods and artificial intelligence methods. Analytical methods are divided into unified and hierarchical solutions based on the scheduling method. Comparative analysis indicates that, compared to the unified solution, the hierarchical solution maintains the independence of each subsystem and achieves a globally optimal solution. However, the repeated iterations required during solving reduce solving efficiency. Artificial intelligence algorithms are primarily divided into methods based on group optimization problems and machine learning algorithms. Although both achieve global optimization, machine learning algorithms demonstrate higher solution rates and robustness compared to methods based on group optimization problems, making them more commonly applied solution algorithms. However, the long offline training time for machine learning algorithms requires further optimization. Finally, the existing problems of demand response mechanisms and their future potential trends are summarized, and an outlook on the participation of demand response in the optimal dispatching of electric and thermal integrated energy systems is provided. This aims to provide a reference for future research on the optimal dispatching of electric and thermal integrated energy systems based on demand response.关键词：demand response;integrated energy systems;flexibility;optimized dispatching;new energy consumption953|455|0更新时间：2025-01-17 - “智能轨道列车动态无线供电系统研究取得新进展，专家基于无源控制理论提出恒功率控制策略，有效提高系统稳定性和可靠性。”摘要：Mutual inductance changes cause fluctuations in the output power of the dynamic wireless power transfer (DWPT) system for autonomous rail trains, affecting the stability and reliability of the system. Traditional linear control strategies exhibit poor dynamic performance and fail to adapt effectively to dynamic wireless power transfer systems. Based on passive control theory, this study proposes a constant power control strategy suitable for the DWPT system of autonomous rail trains, characterized by fast response and strong robustness. The structure of the autonomous-rail train DWPT system is introduced, and the equivalent circuit model of the system is established. The transmission and startup characteristics are analyzed, and the necessary conditions for achieving constant power output of the system are obtained. The EL equation of the DC–DC converter is derived to verify the passivity of the converter. A passive-based controller is designed through energy shaping and virtual damping injection, enabling the system output power to remain constant by controlling the expected value of the current flowing into the secondary DC–DC converter. In addition, a pre-charging circuit is designed to protect the electrical equipment, utilizing the onboard energy storage device to reverse charge the energy storage elements in the DC–DC converter, preventing excessive transient impulse current during startup. A corresponding simulation model is built in MATLAB/Simulink, and a 300 kW DWPT system experimental platform for autonomous rail trains is established for testing. The simulation and experimental results showed that, compared to traditional control strategies, the proposed passive-based control strategy enables the DWPT system’s output power to remain unaffected by mutual inductance fluctuations, maintaining an average power of 298.2 kW. In addition, the pre-charging circuit effectively suppresses inrush currents during startup. These results confirm the validity and correctness of the proposed control strategy, highlighting its advantages, including fast response speed and strong robustness.关键词：Autonomous-rail Train;Dynamic Wireless Power Transfer;Passive-Based Control;Constant power output312|286|0更新时间：2025-01-17
- “跨区域能源交互研究取得新进展，专家提出基于连续锥规划方法的多区域电气互联系统分布式协调调度模型，为建立综合能源系统协调调度模型提供参考。”摘要：Multi-area energy exchange addresses the reverse distribution of energy resources and demand in various regions of China through resource and risk sharing. The complementarity and mutual support between different energy sources effectively alleviate the imbalance between the pace of wind power installation and its consumption level. Based on this, this study proposes a distributed coordinated dispatch model for multi-area integrated electricity and natural gas systems (IEGS) using sequential cone programming (SCP) within the context of integrated energy systems. Firstly, considering the economic benefits of day-ahead dispatch, an IEGS economic dispatch model is established, targeting the minimization of total energy consumption costs. Secondly, given the positive impact of the flexible power adjustment characteristics of DC tie-lines on renewable energy consumption, DC tie-lines and pipelines are utilized as carriers for energy flow transmission and resource sharing between regions and are subjected to modeling and analysis. In addition, a second-order cone relaxation method based on SCP is proposed to handle the gas network flow in the IEGS economic dispatch model, linearizing the flow constraints and reducing the relaxation gap caused by optimization processing. Finally, the analytical target cascading (ATC) method is utilized to demonstrate the autonomous capacity of different areas to solve the distributed dispatch model after decoupling shared variables between areas and establishing the distributed coordination dispatch. Under the circumstances of two-region and three-region interconnected IEGS, the impact of different interconnection methods and gas network flow processing methods on dispatch results is verified, and the difference between centralized and distributed algorithms in handling multi-area integrated problems is compared. The simulation results indicated that the second-order cone relaxation based on SCP significantly ameliorates the relaxation gap. The economic efficiency of power grid operation is improved, and the wind curtailment rate is decreased due to DC tie-lines and pipelines. The solution of the distributed algorithm based on ATC is close to the global optimal solution of the distributed algorithm. The distributed coordinated dispatch model proposed in this study is reasonably designed and provides a reference for establishing a coordinated dispatch model for integrated energy systems considering integrating new energy.关键词：integrated energy system;sequential cone programming;DC tie-line;distributed coordinated dispatch242|400|0更新时间：2025-01-17
ELECTRICAL ENGINEERING
- “在数控工艺决策领域，专家提出了基于相似实例重用的铣削类零件数控工艺决策方法，有效降低生产成本，缩短研制周期，提升企业数控工艺设计效率和智能化水平。”摘要：Objective With the rapid development of Intelligent Manufacturing, it has become a research hotspot in both industry and academia. Specifically, intelligent NC process decision-making is an essential supporting method for raising Intelligent Manufacturing, as it focuses on optimizing and deciding the NC process scheme for a part automatically and effectively. Since CAD/CAPP/CAM technologies have been widely applied in modern manufacturing environments, abundant CAD models associated with NC process results have been generated and stored in repositories, implicitly containing engineers’ valuable knowledge and experience. As a result, if the existing NC process achievements can be fully reused, the efficiency and accuracy of NC process decision-making will improve significantly. In contrast, current NC process reuse methods focus on the feature/subpart levels, indicating that research on NC process decision-making at the part level remains limited. Thus, the NC process reuse methods cannot be implemented directly on the shop floor, leading to significant time and labor waste. In addition, the existing NC process decision methods, such as the genetic algorithm, ant colony algorithm, and particle swarm algorithm, have not been combined with the reuse approach. The NC process determination for features still depends on engineers’ interactions. The diversity of NC processes is ignored, as the dynamic changes of manufacturing resources on the shop floor are not considered. Therefore, the flexibility and efficiency are relatively low and cannot meet the requirements for product innovation in manufacturing environments. A similar instance reuse-based NC process decision method for prismatic parts is proposed to overcome shortcomings.Methods Firstly, several concepts are introduced, including operation and process scheme digraph (PSD). The operation is defined as a basic unit consisting of the NC process and involves various characteristics of features. The PSD is a directed graph that describes the precedence of operations for a part. Then, the framework of the proposed approach is illustrated, which explains four significant steps: similar subpart instances retrieval, optimal target function construction, PSD establishment, and NC process scheme optimization. Then, the mathematical model of the evaluation criterion, which considers manufacturing costs, is established. The usage cost of machines and tools and the alteration cost of machines, tools, and fixtures are considered, and the corresponding mathematical functions are provided. The total manufacturing cost is calculated as the weighted sum of the five types of cost. The adaptive generation of multiple NC processes for a new subpart is presented. The previous 3D model retrieval method retrieves and returns similar subpart instances with their NC processes. Considering the dynamic changes in shop floor resources, the detailed parameters of the NC process are modified adaptively, including operation type, processing stage, machine, tool, and tool approach direction. Specifically, flexible candidates for machines, tools, and tool approach directions are added to raise the flexibility of manufacturing. The similarity of each retrieval result is employed to determine the probability of being selected during the optimization iteration with the roulette wheel method. The flexible elements are chosen randomly except for the fixed operation type and processing stage. Once the NC process for each subpart of the new part is determined, the operations are mapped as the vertices of the PSD. The constraint rules of operation precedence are provided to demonstrate the execution priority of operations. Thus, the directed edges of the PSD are obtained, and the entire PSD is determined. Therefore, the optimization objective of the proposed approach is to search the PSD for the shortest path, which corresponds to the optimal NC process scheme with the minimum manufacturing cost. The ant colony and simulated annealing algorithms are combined to solve the problem due to their strong ability for graph searching and global convergence, respectively. Two concepts of graph theory are explained. Then, the basic ant colony parameters are calculated for the optimization objective. The search principle is illustrated: only the vertices with an in-degree equal to 0 can be regarded as the next destination to be visited, which means that the prior operations have been implemented and the precedence is satisfied. The PSD-based ant colony algorithm is proposed and described in detail to generate several locally optimal paths. Then, the paths are input into the simulated annealing algorithm to overcome local convergence and search for the globally optimal NC process scheme.Results and Discussions In the experiments, a subpart consisting of 3 features is utilized to describe the feasibility of adaptive generation of multiple NC processes. A test part with the MBD model is adopted to verify the approach, which is made up of 8 subparts. The state of manufacturing resources is listed, and the precedence constraints of features’ operations are analyzed. When the iteration times reach 100, the global optimal solution is obtained with the minimum manufacturing cost. The machining simulation of the global optimal NC process scheme is conducted with CATIA. The simulation result shows no overcut or interference occurs, and the dimensional accuracy satisfies the quality requirements of the part. Detailed analysis of tool paths for a subpart is implemented to confirm the approach’s effectiveness. The proposed approach’s decision efficiency and time consumption and the typical NC process decision method with CAPP software are compared. The proposed approach requires fewer than 150 human interactions, while the typical method needs more than 1 200 interactions to generate the NC process for subparts. This is because the NC process candidates are obtained adaptively through reusing similar instances. The NC process decision efficiency is improved by 84.6%. On the other hand, the manufacturing cost of the optimal NC process scheme is 16.6% lower.Conclusions The experimental results showed that the proposed approach can generate optimal NC process schemes for parts effectively and automatically, decrease production costs, and shorten the development cycle. It helps to raise the efficiency and intelligence level of NC process decisions for manufacturing environments.关键词：numerical control process decision;reuse;subpart;process scheme digraph193|219|0更新时间：2025-01-17
- “在重离子同步加速器领域，专家研究了超高真空系统的实现方法，通过降低束流管道材料的比放气率和增大真空泵组的抽速，为提高束流寿命提供解决方案。”摘要：During the operation of a heavy-ion synchrotron, an ultra-high vacuum is required to reduce the likelihood of collisions between beam particles and residual gas molecules, increasing the beam’s lifetime. The specific requirement is for the average vacuum in the pipeline to reach an ultra-high vacuum (～10–10 Pa) and for the vacuum throughout the pipeline to exceed the designed value. This study investigates the implementation method of an ultra-high vacuum system for a heavy-ion synchrotron. The vacuum level throughout the pipeline can be achieved even if the vacuum pumps are not distributed uniformly. The main technical approaches are to reduce the pipe material’s outgassing rate and increase the pumping speed of the vacuum pump units. The outgassing rate of the pipe material is reduced by plating a non-evaporable getter (NEG) film on the inner surface of the vacuum pipe and baking the pipe at high temperatures. The pumping speed is increased using a high-pumping-speed titanium sublimation pump (TSP). The vacuum level throughout the pipeline is improved through NEG plating. This method is applied in the vacuum system design of the heavy-ion synchrotron for the upgrade project at the Xi’an 200 MeV Proton Application Facility (XiPAF). The baking solutions include heating jacket baking and high-temperature baking in vacuum ovens. The purpose of vacuum pipe baking is to reduce the specific outgassing rate of the inner surface of the vacuum pipe. The heating jacket is mounted on the outer surface of the vacuum piping, the titanium sublimation pump housing, and the vacuum chamber. This setup activates the NEG film and removes surface gas from the piping without NEG film coating. The inside surface of the vacuum piping is coated with a NEG film, which has a suction rate of 0.5 L/(s·cm–2) There are 36 titanium sublimation pumps installed in the heavy-ion synchrotron. At the primary vacuum pipe connection sites, a single titanium sublimation pump provides an effective pumping rate of 2200 L/s for hydrogen gas. The Molflow+ software is applied to simulate the vacuum distribution inside the XiPAF heavy-ion synchrotron vacuum piping. Through design optimization, calculations indicated that both the average vacuum degree and the vacuum degree along the annular pipe are expected to exceed 5×10–10 Pa, which meets the physical design requirements of the synchrotron.关键词：Heavy-ion synchrotron;ultra-high vacuum system;Titanium sublimation Pump;Out-gassing rate;NEG film159|517|0更新时间：2025-01-17
- “在煤矿智能化领域，专家提出了新型双阀芯增量式高水基数字阀结构，有效解决了综采工作面直线度自适应控制难题，为煤矿智能化建设提供了有力支撑。”摘要：The straightness control of the scraper conveyor is a key technology for achieving intelligent coal mining. However, the discontinuity of the on-off hydraulic support electro-hydraulic valve in the underground working face makes it difficult to meet the requirements of adaptive straightness control in the integrated mining work surface. In addition, the proportional valve employing an oil medium cannot be directly applied to the hydraulic support system that adopts emulsion as the medium. In view of these problems, a novel incremental high-water-based digital valve with a dual spool is proposed, primarily comprising a valve block, dual valve core components, and a motor screw drive element. Firstly, the structure and operation of the dual-core incremental digital valve are investigated. It operates in both open-loop and closed-loop modes and separately regulates the load port due to its unique double-core arrangement. Secondly, the critical sealing coupling at the valve core and seat is modeled. When matching “hard-to-soft” materials, considering sealing specific pressure and sealing contact width comprehensively, it is proven that the valve seat’s half-cone angle should be 0.61 rad to achieve optimal performance in terms of sealing performance and the service life of the sealing pair. A simulation of the electro-mechanical converter is conducted using sliding mode control, and the simulation results indicate that the sliding mode control algorithm stably follows the reference signal in 0.08 s, performing significantly better in response performance than PID. Finally, the xPC Target real-time control platform is created for prototype testing to verify the sealing pair’s effectiveness and operational performance under various motor subdivisions. The experimental analysis demonstrates that, under 21 MPa, the seal is completely sealed. The subdivision control performance at 1 600 pulses per rotation is optimal. During the 0.2 mm step response, the flow rate reaches 13 L/min with a response time of 0.25 s and a steady-state fault of 3 μm, which performs well.关键词：hydraulic support;high water-based;incremental digital valve;sliding mode control;subdivision control62|132|1更新时间：2025-01-17
MECHANICAL ENGINEERING
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0